{
  "status": "complete",
  "created_at_utc": "2026-09-13T07:39:48.930675+00:00",
  "protocol": {
    "created_before_final_evaluation": true,
    "inputs": "Fresh 64x36 depth image, continuous current XYZ, heading and target XYZ. No NAV IDs, route geometry, teacher or action cache inside deployed policy.",
    "versions": {
      "v6": "Supervised online sensory interface; frozen V3 connectome and V2 PFL3 decoder.",
      "v7": "V6 refined with DAgger: teacher labels at the policy own visited states; same inference architecture.",
      "v8": "PPO from scratch for sensory and motor adapters, original fixed anatomical weights. No V2/V3/V6/V7 learned weights transferred."
    },
    "surface": "Same triangulated replica and physical limits for V6\u2013V8",
    "timeout_seconds": 120,
    "success_xy_m": 2,
    "success_z_m": 0.6,
    "max_speed_m_s": 4.5,
    "max_turn_rad_s": 2.8,
    "neural_step_seconds": 0.01,
    "action_step_seconds": 0.1,
    "training_teacher": "V6/V7 training only; V8 may use geodesic progress as a training reward but no expert action labels.",
    "selection": "Development only, then freeze all selected checkpoints/code before 100 paired final cases. Retain failures.",
    "tests": {
      "primary": "100 continuous start/goal/heading situations on the same known map",
      "controls": "Matched frozen-policy probes with image removed, changed target, and recurrence disabled; no claim of superiority of biological topology."
    },
    "split_hashes": {
      "development": "b6d55dd901ead18007e0d85fa2f0b9691bd591744e932e4afe4d8960f24e1be2",
      "heldout": "9ff459ffeffb369e05b19a2a2c5162828a4f90621e04f89f0a2818e7368f525a",
      "rl-training": "91236125406f2be79614405b73bac513b9ce005ab322ab42f5c32b229fa4ed47"
    }
  },
  "results": {
    "v6": {
      "trials": 100,
      "successes": 40,
      "mean_success_seconds": 23.470000000000002,
      "mean_capped_seconds": 81.388,
      "failures": [
        {
          "id": "v6-heldout-000",
          "duration": 120.0,
          "final_distance": 69.13514441279185,
          "blocked_seconds": 92.6
        },
        {
          "id": "v6-heldout-001",
          "duration": 120.0,
          "final_distance": 13.720300260666322,
          "blocked_seconds": 0.0
        },
        {
          "id": "v6-heldout-004",
          "duration": 120.0,
          "final_distance": 6.831238800381149,
          "blocked_seconds": 94.0
        },
        {
          "id": "v6-heldout-005",
          "duration": 120.0,
          "final_distance": 12.807793393257539,
          "blocked_seconds": 111.6
        },
        {
          "id": "v6-heldout-008",
          "duration": 120.0,
          "final_distance": 31.52367524163266,
          "blocked_seconds": 86.8
        },
        {
          "id": "v6-heldout-010",
          "duration": 120.0,
          "final_distance": 30.987253212530376,
          "blocked_seconds": 92.7
        },
        {
          "id": "v6-heldout-011",
          "duration": 120.0,
          "final_distance": 70.04646426839909,
          "blocked_seconds": 115.8
        },
        {
          "id": "v6-heldout-013",
          "duration": 120.0,
          "final_distance": 59.07203441315437,
          "blocked_seconds": 117.4
        },
        {
          "id": "v6-heldout-016",
          "duration": 120.0,
          "final_distance": 14.351146314429673,
          "blocked_seconds": 92.1
        },
        {
          "id": "v6-heldout-017",
          "duration": 120.0,
          "final_distance": 70.10527527709635,
          "blocked_seconds": 98.1
        },
        {
          "id": "v6-heldout-019",
          "duration": 120.0,
          "final_distance": 9.813307765162408,
          "blocked_seconds": 85.6
        },
        {
          "id": "v6-heldout-020",
          "duration": 120.0,
          "final_distance": 2.5190657045363873,
          "blocked_seconds": 0.0
        },
        {
          "id": "v6-heldout-021",
          "duration": 120.0,
          "final_distance": 3.3756008536978666,
          "blocked_seconds": 37.5
        },
        {
          "id": "v6-heldout-023",
          "duration": 120.0,
          "final_distance": 3.025523143801096,
          "blocked_seconds": 118.9
        },
        {
          "id": "v6-heldout-024",
          "duration": 120.0,
          "final_distance": 70.59990493581728,
          "blocked_seconds": 103.3
        },
        {
          "id": "v6-heldout-025",
          "duration": 120.0,
          "final_distance": 56.523287840549735,
          "blocked_seconds": 118.3
        },
        {
          "id": "v6-heldout-027",
          "duration": 120.0,
          "final_distance": 32.43361524059015,
          "blocked_seconds": 116.5
        },
        {
          "id": "v6-heldout-028",
          "duration": 120.0,
          "final_distance": 3.855700871160387,
          "blocked_seconds": 0.0
        },
        {
          "id": "v6-heldout-033",
          "duration": 120.0,
          "final_distance": 57.51640195336933,
          "blocked_seconds": 119.9
        },
        {
          "id": "v6-heldout-035",
          "duration": 120.0,
          "final_distance": 10.747967086144701,
          "blocked_seconds": 109.4
        },
        {
          "id": "v6-heldout-036",
          "duration": 120.0,
          "final_distance": 19.610998703081112,
          "blocked_seconds": 85.6
        },
        {
          "id": "v6-heldout-037",
          "duration": 120.0,
          "final_distance": 6.641522323593361,
          "blocked_seconds": 0.0
        },
        {
          "id": "v6-heldout-038",
          "duration": 120.0,
          "final_distance": 47.56605858045521,
          "blocked_seconds": 94.2
        },
        {
          "id": "v6-heldout-040",
          "duration": 120.0,
          "final_distance": 10.36156926165026,
          "blocked_seconds": 95.1
        },
        {
          "id": "v6-heldout-042",
          "duration": 120.0,
          "final_distance": 3.191429871405291,
          "blocked_seconds": 0.0
        },
        {
          "id": "v6-heldout-043",
          "duration": 120.0,
          "final_distance": 4.763261888607365,
          "blocked_seconds": 0.0
        },
        {
          "id": "v6-heldout-044",
          "duration": 120.0,
          "final_distance": 19.37993169160348,
          "blocked_seconds": 83.9
        },
        {
          "id": "v6-heldout-048",
          "duration": 120.0,
          "final_distance": 16.06192845888229,
          "blocked_seconds": 84.2
        },
        {
          "id": "v6-heldout-051",
          "duration": 120.0,
          "final_distance": 8.706634848461512,
          "blocked_seconds": 82.6
        },
        {
          "id": "v6-heldout-052",
          "duration": 120.0,
          "final_distance": 80.24202308640486,
          "blocked_seconds": 96.9
        },
        {
          "id": "v6-heldout-053",
          "duration": 120.0,
          "final_distance": 2.90551759339655,
          "blocked_seconds": 119.0
        },
        {
          "id": "v6-heldout-054",
          "duration": 120.0,
          "final_distance": 14.591783308814582,
          "blocked_seconds": 0.0
        },
        {
          "id": "v6-heldout-056",
          "duration": 120.0,
          "final_distance": 18.264489375301743,
          "blocked_seconds": 93.7
        },
        {
          "id": "v6-heldout-057",
          "duration": 120.0,
          "final_distance": 31.692210253198855,
          "blocked_seconds": 108.7
        },
        {
          "id": "v6-heldout-060",
          "duration": 120.0,
          "final_distance": 69.46618209952956,
          "blocked_seconds": 103.6
        },
        {
          "id": "v6-heldout-061",
          "duration": 120.0,
          "final_distance": 23.010278471851873,
          "blocked_seconds": 84.2
        },
        {
          "id": "v6-heldout-062",
          "duration": 120.0,
          "final_distance": 15.458372150659043,
          "blocked_seconds": 0.0
        },
        {
          "id": "v6-heldout-063",
          "duration": 120.0,
          "final_distance": 14.893152880377254,
          "blocked_seconds": 100.6
        },
        {
          "id": "v6-heldout-065",
          "duration": 120.0,
          "final_distance": 37.231429753061924,
          "blocked_seconds": 105.9
        },
        {
          "id": "v6-heldout-067",
          "duration": 120.0,
          "final_distance": 19.67433167167951,
          "blocked_seconds": 93.1
        },
        {
          "id": "v6-heldout-069",
          "duration": 120.0,
          "final_distance": 31.838401050644304,
          "blocked_seconds": 99.5
        },
        {
          "id": "v6-heldout-070",
          "duration": 120.0,
          "final_distance": 66.32963237192922,
          "blocked_seconds": 105.2
        },
        {
          "id": "v6-heldout-071",
          "duration": 120.0,
          "final_distance": 26.883250129702812,
          "blocked_seconds": 102.0
        },
        {
          "id": "v6-heldout-072",
          "duration": 120.0,
          "final_distance": 39.14930083377633,
          "blocked_seconds": 95.7
        },
        {
          "id": "v6-heldout-073",
          "duration": 120.0,
          "final_distance": 9.92486842409773,
          "blocked_seconds": 115.1
        },
        {
          "id": "v6-heldout-076",
          "duration": 120.0,
          "final_distance": 19.294023524734484,
          "blocked_seconds": 100.7
        },
        {
          "id": "v6-heldout-077",
          "duration": 120.0,
          "final_distance": 21.254887181126886,
          "blocked_seconds": 99.2
        },
        {
          "id": "v6-heldout-080",
          "duration": 120.0,
          "final_distance": 2.9940422491551653,
          "blocked_seconds": 97.1
        },
        {
          "id": "v6-heldout-081",
          "duration": 120.0,
          "final_distance": 6.112010261379384,
          "blocked_seconds": 118.5
        },
        {
          "id": "v6-heldout-082",
          "duration": 120.0,
          "final_distance": 33.92576015816317,
          "blocked_seconds": 90.5
        },
        {
          "id": "v6-heldout-083",
          "duration": 120.0,
          "final_distance": 18.321810372012703,
          "blocked_seconds": 94.0
        },
        {
          "id": "v6-heldout-084",
          "duration": 120.0,
          "final_distance": 8.345996930129372,
          "blocked_seconds": 101.6
        },
        {
          "id": "v6-heldout-085",
          "duration": 120.0,
          "final_distance": 60.45244377094422,
          "blocked_seconds": 109.5
        },
        {
          "id": "v6-heldout-088",
          "duration": 120.0,
          "final_distance": 85.76057459928707,
          "blocked_seconds": 104.2
        },
        {
          "id": "v6-heldout-090",
          "duration": 120.0,
          "final_distance": 19.664372897595502,
          "blocked_seconds": 91.5
        },
        {
          "id": "v6-heldout-091",
          "duration": 120.0,
          "final_distance": 2.7952682676224456,
          "blocked_seconds": 91.9
        },
        {
          "id": "v6-heldout-092",
          "duration": 120.0,
          "final_distance": 57.46121375720554,
          "blocked_seconds": 89.3
        },
        {
          "id": "v6-heldout-093",
          "duration": 120.0,
          "final_distance": 96.29858803610189,
          "blocked_seconds": 116.4
        },
        {
          "id": "v6-heldout-095",
          "duration": 120.0,
          "final_distance": 93.41442499409916,
          "blocked_seconds": 117.7
        },
        {
          "id": "v6-heldout-097",
          "duration": 120.0,
          "final_distance": 2.1520102741105505,
          "blocked_seconds": 0.0
        }
      ]
    },
    "v7": {
      "trials": 100,
      "successes": 58,
      "mean_success_seconds": 27.112068965517246,
      "mean_capped_seconds": 66.12499999999999,
      "failures": [
        {
          "id": "v7-heldout-005",
          "duration": 120.0,
          "final_distance": 12.781924695943577,
          "blocked_seconds": 111.9
        },
        {
          "id": "v7-heldout-011",
          "duration": 120.0,
          "final_distance": 75.36268434024262,
          "blocked_seconds": 112.1
        },
        {
          "id": "v7-heldout-016",
          "duration": 120.0,
          "final_distance": 14.29900232794749,
          "blocked_seconds": 93.7
        },
        {
          "id": "v7-heldout-017",
          "duration": 120.0,
          "final_distance": 70.53596490158156,
          "blocked_seconds": 99.4
        },
        {
          "id": "v7-heldout-021",
          "duration": 120.0,
          "final_distance": 58.251736637677766,
          "blocked_seconds": 106.0
        },
        {
          "id": "v7-heldout-023",
          "duration": 120.0,
          "final_distance": 2.387695771735485,
          "blocked_seconds": 118.6
        },
        {
          "id": "v7-heldout-024",
          "duration": 120.0,
          "final_distance": 70.57986408102418,
          "blocked_seconds": 104.3
        },
        {
          "id": "v7-heldout-025",
          "duration": 120.0,
          "final_distance": 20.197782183339974,
          "blocked_seconds": 82.6
        },
        {
          "id": "v7-heldout-029",
          "duration": 120.0,
          "final_distance": 49.123878607643896,
          "blocked_seconds": 97.8
        },
        {
          "id": "v7-heldout-033",
          "duration": 120.0,
          "final_distance": 57.56628632246381,
          "blocked_seconds": 119.8
        },
        {
          "id": "v7-heldout-035",
          "duration": 120.0,
          "final_distance": 4.709409795327051,
          "blocked_seconds": 0.0
        },
        {
          "id": "v7-heldout-036",
          "duration": 120.0,
          "final_distance": 29.573700079571985,
          "blocked_seconds": 92.3
        },
        {
          "id": "v7-heldout-037",
          "duration": 120.0,
          "final_distance": 5.9345882097001565,
          "blocked_seconds": 0.0
        },
        {
          "id": "v7-heldout-038",
          "duration": 120.0,
          "final_distance": 47.355787175568544,
          "blocked_seconds": 94.0
        },
        {
          "id": "v7-heldout-040",
          "duration": 120.0,
          "final_distance": 10.163815492317143,
          "blocked_seconds": 96.9
        },
        {
          "id": "v7-heldout-043",
          "duration": 120.0,
          "final_distance": 5.533428839252477,
          "blocked_seconds": 0.0
        },
        {
          "id": "v7-heldout-051",
          "duration": 120.0,
          "final_distance": 9.053490140918155,
          "blocked_seconds": 16.7
        },
        {
          "id": "v7-heldout-053",
          "duration": 120.0,
          "final_distance": 2.88577578491015,
          "blocked_seconds": 119.0
        },
        {
          "id": "v7-heldout-056",
          "duration": 120.0,
          "final_distance": 18.25317277214579,
          "blocked_seconds": 96.3
        },
        {
          "id": "v7-heldout-057",
          "duration": 120.0,
          "final_distance": 4.4302535786538275,
          "blocked_seconds": 96.4
        },
        {
          "id": "v7-heldout-060",
          "duration": 120.0,
          "final_distance": 69.08864592401683,
          "blocked_seconds": 104.2
        },
        {
          "id": "v7-heldout-061",
          "duration": 120.0,
          "final_distance": 13.677517722561472,
          "blocked_seconds": 77.1
        },
        {
          "id": "v7-heldout-063",
          "duration": 120.0,
          "final_distance": 14.781349065126124,
          "blocked_seconds": 100.0
        },
        {
          "id": "v7-heldout-065",
          "duration": 120.0,
          "final_distance": 37.16289060363336,
          "blocked_seconds": 105.6
        },
        {
          "id": "v7-heldout-068",
          "duration": 120.0,
          "final_distance": 1.001500778780907,
          "blocked_seconds": 97.1
        },
        {
          "id": "v7-heldout-069",
          "duration": 120.0,
          "final_distance": 31.520694402244768,
          "blocked_seconds": 98.1
        },
        {
          "id": "v7-heldout-070",
          "duration": 120.0,
          "final_distance": 66.49406407364675,
          "blocked_seconds": 106.2
        },
        {
          "id": "v7-heldout-071",
          "duration": 120.0,
          "final_distance": 70.23927615181897,
          "blocked_seconds": 119.0
        },
        {
          "id": "v7-heldout-073",
          "duration": 120.0,
          "final_distance": 9.145329678240323,
          "blocked_seconds": 115.5
        },
        {
          "id": "v7-heldout-075",
          "duration": 120.0,
          "final_distance": 5.9182732699547556,
          "blocked_seconds": 0.0
        },
        {
          "id": "v7-heldout-077",
          "duration": 120.0,
          "final_distance": 8.50426974996129,
          "blocked_seconds": 91.6
        },
        {
          "id": "v7-heldout-080",
          "duration": 120.0,
          "final_distance": 2.7926894177431993,
          "blocked_seconds": 97.8
        },
        {
          "id": "v7-heldout-081",
          "duration": 120.0,
          "final_distance": 6.157460996251562,
          "blocked_seconds": 108.1
        },
        {
          "id": "v7-heldout-083",
          "duration": 120.0,
          "final_distance": 18.536410067186797,
          "blocked_seconds": 94.2
        },
        {
          "id": "v7-heldout-085",
          "duration": 120.0,
          "final_distance": 60.07756067843124,
          "blocked_seconds": 109.5
        },
        {
          "id": "v7-heldout-088",
          "duration": 120.0,
          "final_distance": 28.9989690521046,
          "blocked_seconds": 73.6
        },
        {
          "id": "v7-heldout-089",
          "duration": 120.0,
          "final_distance": 30.364501906391112,
          "blocked_seconds": 90.5
        },
        {
          "id": "v7-heldout-090",
          "duration": 120.0,
          "final_distance": 19.658927355440177,
          "blocked_seconds": 93.2
        },
        {
          "id": "v7-heldout-091",
          "duration": 120.0,
          "final_distance": 21.507177189309907,
          "blocked_seconds": 104.4
        },
        {
          "id": "v7-heldout-092",
          "duration": 120.0,
          "final_distance": 85.102432490076,
          "blocked_seconds": 104.8
        },
        {
          "id": "v7-heldout-093",
          "duration": 120.0,
          "final_distance": 100.58560623059358,
          "blocked_seconds": 118.7
        },
        {
          "id": "v7-heldout-095",
          "duration": 120.0,
          "final_distance": 92.05694356413902,
          "blocked_seconds": 118.4
        }
      ]
    },
    "v8": {
      "trials": 100,
      "successes": 16,
      "mean_success_seconds": 6.59375,
      "mean_capped_seconds": 101.855,
      "failures": [
        {
          "id": "v8-heldout-000",
          "duration": 120.0,
          "final_distance": 99.13281612330266,
          "blocked_seconds": 114.8
        },
        {
          "id": "v8-heldout-001",
          "duration": 120.0,
          "final_distance": 8.66661874168429,
          "blocked_seconds": 109.3
        },
        {
          "id": "v8-heldout-003",
          "duration": 120.0,
          "final_distance": 90.4028290250603,
          "blocked_seconds": 119.4
        },
        {
          "id": "v8-heldout-004",
          "duration": 120.0,
          "final_distance": 26.15523789564516,
          "blocked_seconds": 110.3
        },
        {
          "id": "v8-heldout-005",
          "duration": 120.0,
          "final_distance": 16.809006355666117,
          "blocked_seconds": 120.0
        },
        {
          "id": "v8-heldout-007",
          "duration": 120.0,
          "final_distance": 4.028303401510728,
          "blocked_seconds": 119.7
        },
        {
          "id": "v8-heldout-008",
          "duration": 120.0,
          "final_distance": 86.21369157613405,
          "blocked_seconds": 112.3
        },
        {
          "id": "v8-heldout-010",
          "duration": 120.0,
          "final_distance": 57.433926828788316,
          "blocked_seconds": 109.7
        },
        {
          "id": "v8-heldout-011",
          "duration": 120.0,
          "final_distance": 59.17094564803027,
          "blocked_seconds": 119.5
        },
        {
          "id": "v8-heldout-013",
          "duration": 120.0,
          "final_distance": 58.85654119781424,
          "blocked_seconds": 118.6
        },
        {
          "id": "v8-heldout-015",
          "duration": 120.0,
          "final_distance": 60.5650746576632,
          "blocked_seconds": 119.6
        },
        {
          "id": "v8-heldout-016",
          "duration": 120.0,
          "final_distance": 59.41625966620737,
          "blocked_seconds": 114.6
        },
        {
          "id": "v8-heldout-017",
          "duration": 120.0,
          "final_distance": 83.4103787098623,
          "blocked_seconds": 113.6
        },
        {
          "id": "v8-heldout-019",
          "duration": 120.0,
          "final_distance": 81.81620636948327,
          "blocked_seconds": 115.7
        },
        {
          "id": "v8-heldout-020",
          "duration": 120.0,
          "final_distance": 32.21789828201226,
          "blocked_seconds": 118.5
        },
        {
          "id": "v8-heldout-021",
          "duration": 120.0,
          "final_distance": 72.50341051355295,
          "blocked_seconds": 117.4
        },
        {
          "id": "v8-heldout-023",
          "duration": 120.0,
          "final_distance": 2.01128942059259,
          "blocked_seconds": 118.9
        },
        {
          "id": "v8-heldout-024",
          "duration": 120.0,
          "final_distance": 81.28551732663179,
          "blocked_seconds": 117.2
        },
        {
          "id": "v8-heldout-025",
          "duration": 120.0,
          "final_distance": 56.87575886667307,
          "blocked_seconds": 119.9
        },
        {
          "id": "v8-heldout-026",
          "duration": 120.0,
          "final_distance": 26.22532005159725,
          "blocked_seconds": 106.8
        },
        {
          "id": "v8-heldout-027",
          "duration": 120.0,
          "final_distance": 33.29740233274034,
          "blocked_seconds": 119.4
        },
        {
          "id": "v8-heldout-029",
          "duration": 120.0,
          "final_distance": 49.16299358847636,
          "blocked_seconds": 120.0
        },
        {
          "id": "v8-heldout-030",
          "duration": 120.0,
          "final_distance": 102.52239850465077,
          "blocked_seconds": 115.6
        },
        {
          "id": "v8-heldout-031",
          "duration": 120.0,
          "final_distance": 58.78351885592223,
          "blocked_seconds": 115.7
        },
        {
          "id": "v8-heldout-033",
          "duration": 120.0,
          "final_distance": 57.477904015437566,
          "blocked_seconds": 120.0
        },
        {
          "id": "v8-heldout-035",
          "duration": 120.0,
          "final_distance": 2.109827033577383,
          "blocked_seconds": 112.8
        },
        {
          "id": "v8-heldout-036",
          "duration": 120.0,
          "final_distance": 65.28790859990858,
          "blocked_seconds": 116.3
        },
        {
          "id": "v8-heldout-037",
          "duration": 120.0,
          "final_distance": 49.3945614405599,
          "blocked_seconds": 114.4
        },
        {
          "id": "v8-heldout-038",
          "duration": 120.0,
          "final_distance": 93.25539093410599,
          "blocked_seconds": 115.8
        },
        {
          "id": "v8-heldout-039",
          "duration": 120.0,
          "final_distance": 45.55126694123781,
          "blocked_seconds": 118.4
        },
        {
          "id": "v8-heldout-040",
          "duration": 120.0,
          "final_distance": 44.38445076089334,
          "blocked_seconds": 118.4
        },
        {
          "id": "v8-heldout-041",
          "duration": 120.0,
          "final_distance": 36.09071749538502,
          "blocked_seconds": 117.2
        },
        {
          "id": "v8-heldout-042",
          "duration": 120.0,
          "final_distance": 28.813928751863784,
          "blocked_seconds": 116.8
        },
        {
          "id": "v8-heldout-043",
          "duration": 120.0,
          "final_distance": 44.60153293830858,
          "blocked_seconds": 118.3
        },
        {
          "id": "v8-heldout-044",
          "duration": 120.0,
          "final_distance": 67.42242339051455,
          "blocked_seconds": 117.4
        },
        {
          "id": "v8-heldout-046",
          "duration": 120.0,
          "final_distance": 40.14581147861487,
          "blocked_seconds": 115.0
        },
        {
          "id": "v8-heldout-047",
          "duration": 120.0,
          "final_distance": 55.51899408041302,
          "blocked_seconds": 119.1
        },
        {
          "id": "v8-heldout-048",
          "duration": 120.0,
          "final_distance": 64.62658651480562,
          "blocked_seconds": 116.8
        },
        {
          "id": "v8-heldout-051",
          "duration": 120.0,
          "final_distance": 29.87965494621156,
          "blocked_seconds": 106.2
        },
        {
          "id": "v8-heldout-052",
          "duration": 120.0,
          "final_distance": 97.2697732244423,
          "blocked_seconds": 115.6
        },
        {
          "id": "v8-heldout-053",
          "duration": 120.0,
          "final_distance": 2.270871842444415,
          "blocked_seconds": 119.6
        },
        {
          "id": "v8-heldout-054",
          "duration": 120.0,
          "final_distance": 12.174953232559373,
          "blocked_seconds": 114.6
        },
        {
          "id": "v8-heldout-055",
          "duration": 120.0,
          "final_distance": 69.754116216805,
          "blocked_seconds": 119.5
        },
        {
          "id": "v8-heldout-056",
          "duration": 120.0,
          "final_distance": 38.746397751533756,
          "blocked_seconds": 108.4
        },
        {
          "id": "v8-heldout-057",
          "duration": 120.0,
          "final_distance": 38.72154224124802,
          "blocked_seconds": 119.2
        },
        {
          "id": "v8-heldout-059",
          "duration": 120.0,
          "final_distance": 38.638290939849774,
          "blocked_seconds": 111.1
        },
        {
          "id": "v8-heldout-060",
          "duration": 120.0,
          "final_distance": 79.08342602522579,
          "blocked_seconds": 116.0
        },
        {
          "id": "v8-heldout-061",
          "duration": 120.0,
          "final_distance": 64.91491797839059,
          "blocked_seconds": 119.7
        },
        {
          "id": "v8-heldout-062",
          "duration": 120.0,
          "final_distance": 10.856353711697478,
          "blocked_seconds": 116.5
        },
        {
          "id": "v8-heldout-063",
          "duration": 120.0,
          "final_distance": 37.22160810076232,
          "blocked_seconds": 119.9
        },
        {
          "id": "v8-heldout-064",
          "duration": 120.0,
          "final_distance": 59.16501607675047,
          "blocked_seconds": 118.2
        },
        {
          "id": "v8-heldout-065",
          "duration": 120.0,
          "final_distance": 27.24368892163767,
          "blocked_seconds": 108.1
        },
        {
          "id": "v8-heldout-066",
          "duration": 120.0,
          "final_distance": 100.01295588505714,
          "blocked_seconds": 115.8
        },
        {
          "id": "v8-heldout-067",
          "duration": 120.0,
          "final_distance": 57.45435041005952,
          "blocked_seconds": 119.6
        },
        {
          "id": "v8-heldout-068",
          "duration": 120.0,
          "final_distance": 40.70435508867035,
          "blocked_seconds": 117.1
        },
        {
          "id": "v8-heldout-069",
          "duration": 120.0,
          "final_distance": 60.89398291915246,
          "blocked_seconds": 120.0
        },
        {
          "id": "v8-heldout-070",
          "duration": 120.0,
          "final_distance": 79.37364836281097,
          "blocked_seconds": 117.8
        },
        {
          "id": "v8-heldout-071",
          "duration": 120.0,
          "final_distance": 70.68435339391276,
          "blocked_seconds": 119.5
        },
        {
          "id": "v8-heldout-072",
          "duration": 120.0,
          "final_distance": 48.795143023553514,
          "blocked_seconds": 116.8
        },
        {
          "id": "v8-heldout-073",
          "duration": 120.0,
          "final_distance": 12.120159199304753,
          "blocked_seconds": 117.5
        },
        {
          "id": "v8-heldout-074",
          "duration": 120.0,
          "final_distance": 49.18424487896547,
          "blocked_seconds": 113.2
        },
        {
          "id": "v8-heldout-075",
          "duration": 120.0,
          "final_distance": 35.41259472678393,
          "blocked_seconds": 118.7
        },
        {
          "id": "v8-heldout-076",
          "duration": 120.0,
          "final_distance": 48.56332847636782,
          "blocked_seconds": 117.1
        },
        {
          "id": "v8-heldout-077",
          "duration": 120.0,
          "final_distance": 42.915866236833125,
          "blocked_seconds": 116.0
        },
        {
          "id": "v8-heldout-079",
          "duration": 120.0,
          "final_distance": 15.061308010686975,
          "blocked_seconds": 119.7
        },
        {
          "id": "v8-heldout-080",
          "duration": 120.0,
          "final_distance": 43.45047489368449,
          "blocked_seconds": 115.6
        },
        {
          "id": "v8-heldout-081",
          "duration": 120.0,
          "final_distance": 6.4777766209659005,
          "blocked_seconds": 120.0
        },
        {
          "id": "v8-heldout-082",
          "duration": 120.0,
          "final_distance": 87.47878310103658,
          "blocked_seconds": 113.8
        },
        {
          "id": "v8-heldout-083",
          "duration": 120.0,
          "final_distance": 81.30536476931725,
          "blocked_seconds": 119.2
        },
        {
          "id": "v8-heldout-084",
          "duration": 120.0,
          "final_distance": 26.829437767253552,
          "blocked_seconds": 115.8
        },
        {
          "id": "v8-heldout-085",
          "duration": 120.0,
          "final_distance": 68.34320137403809,
          "blocked_seconds": 117.3
        },
        {
          "id": "v8-heldout-086",
          "duration": 120.0,
          "final_distance": 70.14695586906447,
          "blocked_seconds": 116.8
        },
        {
          "id": "v8-heldout-087",
          "duration": 120.0,
          "final_distance": 32.127583042928435,
          "blocked_seconds": 114.6
        },
        {
          "id": "v8-heldout-088",
          "duration": 120.0,
          "final_distance": 93.97730832042146,
          "blocked_seconds": 115.2
        },
        {
          "id": "v8-heldout-089",
          "duration": 120.0,
          "final_distance": 92.62205136175112,
          "blocked_seconds": 120.0
        },
        {
          "id": "v8-heldout-090",
          "duration": 120.0,
          "final_distance": 59.08326915435506,
          "blocked_seconds": 116.1
        },
        {
          "id": "v8-heldout-091",
          "duration": 120.0,
          "final_distance": 28.47037799180925,
          "blocked_seconds": 116.9
        },
        {
          "id": "v8-heldout-092",
          "duration": 120.0,
          "final_distance": 88.21672628705309,
          "blocked_seconds": 112.9
        },
        {
          "id": "v8-heldout-093",
          "duration": 120.0,
          "final_distance": 101.63382697980765,
          "blocked_seconds": 120.0
        },
        {
          "id": "v8-heldout-094",
          "duration": 120.0,
          "final_distance": 40.42084621255549,
          "blocked_seconds": 119.3
        },
        {
          "id": "v8-heldout-095",
          "duration": 120.0,
          "final_distance": 88.76803194451753,
          "blocked_seconds": 119.4
        },
        {
          "id": "v8-heldout-097",
          "duration": 120.0,
          "final_distance": 2.198730804589989,
          "blocked_seconds": 119.9
        },
        {
          "id": "v8-heldout-098",
          "duration": 120.0,
          "final_distance": 21.357098402404503,
          "blocked_seconds": 115.0
        },
        {
          "id": "v8-heldout-099",
          "duration": 120.0,
          "final_distance": 41.9519845506032,
          "blocked_seconds": 119.5
        }
      ]
    }
  },
  "controls": {
    "v6": {
      "normal": {
        "trials": 8,
        "successes": 5,
        "mean_success_seconds": 37.9,
        "mean_capped_seconds": 68.6875,
        "failures": [
          {
            "id": "v6-development-000",
            "duration": 120.0,
            "final_distance": 85.63051285214281,
            "blocked_seconds": 0.0
          },
          {
            "id": "v6-development-002",
            "duration": 120.0,
            "final_distance": 4.107832587639623,
            "blocked_seconds": 66.3
          },
          {
            "id": "v6-development-007",
            "duration": 120.0,
            "final_distance": 46.39701856148144,
            "blocked_seconds": 117.9
          }
        ]
      },
      "no_image": {
        "trials": 8,
        "successes": 0,
        "mean_success_seconds": null,
        "mean_capped_seconds": 120.0,
        "failures": [
          {
            "id": "v6-development-000",
            "duration": 120.0,
            "final_distance": 98.87354013383086,
            "blocked_seconds": 0.0
          },
          {
            "id": "v6-development-001",
            "duration": 120.0,
            "final_distance": 48.84187913428246,
            "blocked_seconds": 111.4
          },
          {
            "id": "v6-development-002",
            "duration": 120.0,
            "final_distance": 76.7060098491379,
            "blocked_seconds": 0.0
          },
          {
            "id": "v6-development-003",
            "duration": 120.0,
            "final_distance": 29.779960412594075,
            "blocked_seconds": 115.0
          },
          {
            "id": "v6-development-004",
            "duration": 120.0,
            "final_distance": 74.65077651042326,
            "blocked_seconds": 10.4
          },
          {
            "id": "v6-development-005",
            "duration": 120.0,
            "final_distance": 105.45111736990359,
            "blocked_seconds": 8.4
          },
          {
            "id": "v6-development-006",
            "duration": 120.0,
            "final_distance": 45.190736000443515,
            "blocked_seconds": 0.0
          },
          {
            "id": "v6-development-007",
            "duration": 120.0,
            "final_distance": 42.23286514513056,
            "blocked_seconds": 0.0
          }
        ]
      },
      "no_recurrence": {
        "trials": 8,
        "successes": 0,
        "mean_success_seconds": null,
        "mean_capped_seconds": 120.0,
        "failures": [
          {
            "id": "v6-development-000",
            "duration": 120.0,
            "final_distance": 108.06287737485015,
            "blocked_seconds": 0.0
          },
          {
            "id": "v6-development-001",
            "duration": 120.0,
            "final_distance": 50.142852623495216,
            "blocked_seconds": 0.0
          },
          {
            "id": "v6-development-002",
            "duration": 120.0,
            "final_distance": 107.90684265460335,
            "blocked_seconds": 0.0
          },
          {
            "id": "v6-development-003",
            "duration": 120.0,
            "final_distance": 26.052604048401747,
            "blocked_seconds": 0.0
          },
          {
            "id": "v6-development-004",
            "duration": 120.0,
            "final_distance": 94.48020186927964,
            "blocked_seconds": 0.0
          },
          {
            "id": "v6-development-005",
            "duration": 120.0,
            "final_distance": 109.19162193837285,
            "blocked_seconds": 0.0
          },
          {
            "id": "v6-development-006",
            "duration": 120.0,
            "final_distance": 56.68932165160241,
            "blocked_seconds": 0.0
          },
          {
            "id": "v6-development-007",
            "duration": 120.0,
            "final_distance": 41.09071113230728,
            "blocked_seconds": 0.0
          }
        ]
      }
    },
    "v7": {
      "normal": {
        "trials": 8,
        "successes": 6,
        "mean_success_seconds": 44.76666666666666,
        "mean_capped_seconds": 63.575,
        "failures": [
          {
            "id": "v7-development-002",
            "duration": 120.0,
            "final_distance": 6.666367558644888,
            "blocked_seconds": 68.6
          },
          {
            "id": "v7-development-004",
            "duration": 120.0,
            "final_distance": 28.32533074611917,
            "blocked_seconds": 81.3
          }
        ]
      },
      "no_image": {
        "trials": 8,
        "successes": 0,
        "mean_success_seconds": null,
        "mean_capped_seconds": 120.0,
        "failures": [
          {
            "id": "v7-development-000",
            "duration": 120.0,
            "final_distance": 96.13980394012752,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-001",
            "duration": 120.0,
            "final_distance": 39.22821673421039,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-002",
            "duration": 120.0,
            "final_distance": 94.06628576320395,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-003",
            "duration": 120.0,
            "final_distance": 29.579091640014536,
            "blocked_seconds": 115.9
          },
          {
            "id": "v7-development-004",
            "duration": 120.0,
            "final_distance": 87.1223435943339,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-005",
            "duration": 120.0,
            "final_distance": 105.38352417379112,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-006",
            "duration": 120.0,
            "final_distance": 46.53034032989635,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-007",
            "duration": 120.0,
            "final_distance": 41.42587708486125,
            "blocked_seconds": 0.0
          }
        ]
      },
      "no_recurrence": {
        "trials": 8,
        "successes": 0,
        "mean_success_seconds": null,
        "mean_capped_seconds": 120.0,
        "failures": [
          {
            "id": "v7-development-000",
            "duration": 120.0,
            "final_distance": 108.06287737485015,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-001",
            "duration": 120.0,
            "final_distance": 50.142852623495216,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-002",
            "duration": 120.0,
            "final_distance": 107.90684265460335,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-003",
            "duration": 120.0,
            "final_distance": 26.052604048401747,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-004",
            "duration": 120.0,
            "final_distance": 94.48020186927964,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-005",
            "duration": 120.0,
            "final_distance": 109.19162193837285,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-006",
            "duration": 120.0,
            "final_distance": 56.68932165160241,
            "blocked_seconds": 0.0
          },
          {
            "id": "v7-development-007",
            "duration": 120.0,
            "final_distance": 41.09071113230728,
            "blocked_seconds": 0.0
          }
        ]
      }
    },
    "v8": {
      "normal": {
        "trials": 8,
        "successes": 1,
        "mean_success_seconds": 21.3,
        "mean_capped_seconds": 107.6625,
        "failures": [
          {
            "id": "v8-development-000",
            "duration": 120.0,
            "final_distance": 114.93025184839975,
            "blocked_seconds": 118.2
          },
          {
            "id": "v8-development-001",
            "duration": 120.0,
            "final_distance": 47.97321775320606,
            "blocked_seconds": 119.5
          },
          {
            "id": "v8-development-002",
            "duration": 120.0,
            "final_distance": 100.9422036734083,
            "blocked_seconds": 115.8
          },
          {
            "id": "v8-development-003",
            "duration": 120.0,
            "final_distance": 23.70116617608795,
            "blocked_seconds": 118.2
          },
          {
            "id": "v8-development-004",
            "duration": 120.0,
            "final_distance": 83.07384528267383,
            "blocked_seconds": 114.5
          },
          {
            "id": "v8-development-005",
            "duration": 120.0,
            "final_distance": 85.78476350671838,
            "blocked_seconds": 111.4
          },
          {
            "id": "v8-development-007",
            "duration": 120.0,
            "final_distance": 40.16739816718311,
            "blocked_seconds": 117.7
          }
        ]
      },
      "no_image": {
        "trials": 8,
        "successes": 0,
        "mean_success_seconds": null,
        "mean_capped_seconds": 120.0,
        "failures": [
          {
            "id": "v8-development-000",
            "duration": 120.0,
            "final_distance": 96.91035896587881,
            "blocked_seconds": 114.5
          },
          {
            "id": "v8-development-001",
            "duration": 120.0,
            "final_distance": 48.084824009541585,
            "blocked_seconds": 119.5
          },
          {
            "id": "v8-development-002",
            "duration": 120.0,
            "final_distance": 99.57139008956317,
            "blocked_seconds": 116.1
          },
          {
            "id": "v8-development-003",
            "duration": 120.0,
            "final_distance": 27.033865096471775,
            "blocked_seconds": 118.6
          },
          {
            "id": "v8-development-004",
            "duration": 120.0,
            "final_distance": 79.08456980611001,
            "blocked_seconds": 114.9
          },
          {
            "id": "v8-development-005",
            "duration": 120.0,
            "final_distance": 103.64606650048006,
            "blocked_seconds": 117.8
          },
          {
            "id": "v8-development-006",
            "duration": 120.0,
            "final_distance": 41.44955044818207,
            "blocked_seconds": 113.4
          },
          {
            "id": "v8-development-007",
            "duration": 120.0,
            "final_distance": 43.692389649100136,
            "blocked_seconds": 118.9
          }
        ]
      },
      "no_recurrence": {
        "trials": 8,
        "successes": 0,
        "mean_success_seconds": null,
        "mean_capped_seconds": 120.0,
        "failures": [
          {
            "id": "v8-development-000",
            "duration": 120.0,
            "final_distance": 108.06287737485015,
            "blocked_seconds": 0.0
          },
          {
            "id": "v8-development-001",
            "duration": 120.0,
            "final_distance": 50.142852623495216,
            "blocked_seconds": 0.0
          },
          {
            "id": "v8-development-002",
            "duration": 120.0,
            "final_distance": 107.90684265460335,
            "blocked_seconds": 0.0
          },
          {
            "id": "v8-development-003",
            "duration": 120.0,
            "final_distance": 26.052604048401747,
            "blocked_seconds": 0.0
          },
          {
            "id": "v8-development-004",
            "duration": 120.0,
            "final_distance": 94.48020186927964,
            "blocked_seconds": 0.0
          },
          {
            "id": "v8-development-005",
            "duration": 120.0,
            "final_distance": 109.19162193837285,
            "blocked_seconds": 0.0
          },
          {
            "id": "v8-development-006",
            "duration": 120.0,
            "final_distance": 56.68932165160241,
            "blocked_seconds": 0.0
          },
          {
            "id": "v8-development-007",
            "duration": 120.0,
            "final_distance": 41.09071113230728,
            "blocked_seconds": 0.0
          }
        ]
      }
    }
  },
  "frozen": {
    "frozen_at_utc": "2026-09-13T07:32:27.399723+00:00",
    "selection": {
      "criterion": "Development success, then capped duration; all final cases untouched until this freeze.",
      "models": {
        "v6": {
          "checkpoint": "outputs/online/checkpoint/v6-broad-w1024-08000.pt",
          "report": "outputs/online/v6-development-broad-w1024-08000.json",
          "width": 1024,
          "steps": 8000,
          "successes": 19,
          "trials": 32,
          "mean_capped_seconds": 67.521875,
          "training_log": "outputs/online/training-v6-broad-w1024.json"
        },
        "v7": {
          "checkpoint": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
          "report": "outputs/online/v7-development-broad-r2-08000.json",
          "width": 1024,
          "round": 2,
          "steps": 8000,
          "successes": 22,
          "trials": 32,
          "mean_capped_seconds": 61.509375,
          "training_log": "outputs/online/training-v7-broad-r2.json",
          "resume": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
          "datasets": [
            "outputs/online/imitation-initial.npz",
            "outputs/online/imitation-broad.npz",
            "outputs/online/dagger-broad-r1.npz",
            "outputs/online/dagger-broad-r2.npz"
          ]
        },
        "v8": {
          "checkpoint": "outputs/online/checkpoint/rl-gpu-100.pt",
          "report": "outputs/online/v8-development-gpu-100.json",
          "updates": 100,
          "environment_steps": 409600,
          "reward_mode": "center",
          "successes": 4,
          "trials": 32,
          "mean_capped_seconds": 106.059375,
          "training_log": "outputs/online/training-rl-gpu.json"
        }
      },
      "studies": {
        "v6": {
          "method": "Highest development success, then lowest capped duration; ties favour smaller width and fewer updates. No final tests used.",
          "candidates": [
            {
              "checkpoint": "outputs/online/checkpoint/v6-broad-w512-08000.pt",
              "report": "outputs/online/v6-development-broad-w512-08000.json",
              "width": 512,
              "steps": 8000,
              "successes": 15,
              "trials": 32,
              "mean_capped_seconds": 77.721875,
              "training_log": "outputs/online/training-v6-broad-w512.json"
            },
            {
              "checkpoint": "outputs/online/checkpoint/v6-broad-w512-16000.pt",
              "report": "outputs/online/v6-development-broad-w512-16000.json",
              "width": 512,
              "steps": 16000,
              "successes": 13,
              "trials": 32,
              "mean_capped_seconds": 83.403125,
              "training_log": "outputs/online/training-v6-broad-w512.json"
            },
            {
              "checkpoint": "outputs/online/checkpoint/v6-broad-w512-24000.pt",
              "report": "outputs/online/v6-development-broad-w512-24000.json",
              "width": 512,
              "steps": 24000,
              "successes": 9,
              "trials": 32,
              "mean_capped_seconds": 94.621875,
              "training_log": "outputs/online/training-v6-broad-w512.json"
            },
            {
              "checkpoint": "outputs/online/checkpoint/v6-broad-w1024-08000.pt",
              "report": "outputs/online/v6-development-broad-w1024-08000.json",
              "width": 1024,
              "steps": 8000,
              "successes": 19,
              "trials": 32,
              "mean_capped_seconds": 67.521875,
              "training_log": "outputs/online/training-v6-broad-w1024.json"
            },
            {
              "checkpoint": "outputs/online/checkpoint/v6-broad-w1024-16000.pt",
              "report": "outputs/online/v6-development-broad-w1024-16000.json",
              "width": 1024,
              "steps": 16000,
              "successes": 15,
              "trials": 32,
              "mean_capped_seconds": 77.215625,
              "training_log": "outputs/online/training-v6-broad-w1024.json"
            },
            {
              "checkpoint": "outputs/online/checkpoint/v6-broad-w1024-24000.pt",
              "report": "outputs/online/v6-development-broad-w1024-24000.json",
              "width": 1024,
              "steps": 24000,
              "successes": 14,
              "trials": 32,
              "mean_capped_seconds": 80.88125,
              "training_log": "outputs/online/training-v6-broad-w1024.json"
            }
          ],
          "selected": {
            "checkpoint": "outputs/online/checkpoint/v6-broad-w1024-08000.pt",
            "report": "outputs/online/v6-development-broad-w1024-08000.json",
            "width": 1024,
            "steps": 8000,
            "successes": 19,
            "trials": 32,
            "mean_capped_seconds": 67.521875,
            "training_log": "outputs/online/training-v6-broad-w1024.json"
          },
          "selected_sha256": "3eb1b01a2bcbbe400d86f4b31360434bc61a50c879379f98c6ef82131b2e07f6"
        },
        "v7": {
          "method": "Highest development success, then lowest capped duration; ties favour earlier round and fewer updates. No final tests used.",
          "v6_start": {
            "checkpoint": "outputs/online/checkpoint/v6-broad-w1024-08000.pt",
            "report": "outputs/online/v6-development-broad-w1024-08000.json",
            "width": 1024,
            "steps": 8000,
            "successes": 19,
            "trials": 32,
            "mean_capped_seconds": 67.521875,
            "training_log": "outputs/online/training-v6-broad-w1024.json"
          },
          "rounds": [
            {
              "iteration": 1,
              "collection_checkpoint": "outputs/online/checkpoint/v6-selected-development.pt",
              "training_entries": 128,
              "training_offset": 0,
              "collection_report": "outputs/online/v7-rl-training-broad-r1.json",
              "selected_for_next_round": {
                "checkpoint": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
                "report": "outputs/online/v7-development-broad-r1-04000.json",
                "width": 1024,
                "round": 1,
                "steps": 4000,
                "successes": 17,
                "trials": 32,
                "mean_capped_seconds": 68.946875,
                "training_log": "outputs/online/training-v7-broad-r1.json",
                "resume": "outputs/online/checkpoint/v6-selected-development.pt",
                "datasets": [
                  "outputs/online/imitation-initial.npz",
                  "outputs/online/imitation-broad.npz",
                  "outputs/online/dagger-broad-r1.npz"
                ]
              }
            },
            {
              "iteration": 2,
              "collection_checkpoint": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
              "training_entries": 128,
              "training_offset": 128,
              "collection_report": "outputs/online/v7-rl-training-broad-r2.json",
              "selected_for_next_round": {
                "checkpoint": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
                "report": "outputs/online/v7-development-broad-r2-08000.json",
                "width": 1024,
                "round": 2,
                "steps": 8000,
                "successes": 22,
                "trials": 32,
                "mean_capped_seconds": 61.509375,
                "training_log": "outputs/online/training-v7-broad-r2.json",
                "resume": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
                "datasets": [
                  "outputs/online/imitation-initial.npz",
                  "outputs/online/imitation-broad.npz",
                  "outputs/online/dagger-broad-r1.npz",
                  "outputs/online/dagger-broad-r2.npz"
                ]
              }
            },
            {
              "iteration": 3,
              "collection_checkpoint": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
              "training_entries": 128,
              "training_offset": 256,
              "collection_report": "outputs/online/v7-rl-training-broad-r3.json",
              "selected_for_next_round": {
                "checkpoint": "outputs/online/checkpoint/v7-broad-r3-04000.pt",
                "report": "outputs/online/v7-development-broad-r3-04000.json",
                "width": 1024,
                "round": 3,
                "steps": 4000,
                "successes": 20,
                "trials": 32,
                "mean_capped_seconds": 60.25625,
                "training_log": "outputs/online/training-v7-broad-r3.json",
                "resume": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
                "datasets": [
                  "outputs/online/imitation-initial.npz",
                  "outputs/online/imitation-broad.npz",
                  "outputs/online/dagger-broad-r1.npz",
                  "outputs/online/dagger-broad-r2.npz",
                  "outputs/online/dagger-broad-r3.npz"
                ]
              }
            }
          ],
          "candidates": [
            {
              "checkpoint": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
              "report": "outputs/online/v7-development-broad-r1-04000.json",
              "width": 1024,
              "round": 1,
              "steps": 4000,
              "successes": 17,
              "trials": 32,
              "mean_capped_seconds": 68.946875,
              "training_log": "outputs/online/training-v7-broad-r1.json",
              "resume": "outputs/online/checkpoint/v6-selected-development.pt",
              "datasets": [
                "outputs/online/imitation-initial.npz",
                "outputs/online/imitation-broad.npz",
                "outputs/online/dagger-broad-r1.npz"
              ]
            },
            {
              "checkpoint": "outputs/online/checkpoint/v7-broad-r1-08000.pt",
              "report": "outputs/online/v7-development-broad-r1-08000.json",
              "width": 1024,
              "round": 1,
              "steps": 8000,
              "successes": 14,
              "trials": 32,
              "mean_capped_seconds": 76.2,
              "training_log": "outputs/online/training-v7-broad-r1.json",
              "resume": "outputs/online/checkpoint/v6-selected-development.pt",
              "datasets": [
                "outputs/online/imitation-initial.npz",
                "outputs/online/imitation-broad.npz",
                "outputs/online/dagger-broad-r1.npz"
              ]
            },
            {
              "checkpoint": "outputs/online/checkpoint/v7-broad-r2-04000.pt",
              "report": "outputs/online/v7-development-broad-r2-04000.json",
              "width": 1024,
              "round": 2,
              "steps": 4000,
              "successes": 20,
              "trials": 32,
              "mean_capped_seconds": 61.696875,
              "training_log": "outputs/online/training-v7-broad-r2.json",
              "resume": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
              "datasets": [
                "outputs/online/imitation-initial.npz",
                "outputs/online/imitation-broad.npz",
                "outputs/online/dagger-broad-r1.npz",
                "outputs/online/dagger-broad-r2.npz"
              ]
            },
            {
              "checkpoint": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
              "report": "outputs/online/v7-development-broad-r2-08000.json",
              "width": 1024,
              "round": 2,
              "steps": 8000,
              "successes": 22,
              "trials": 32,
              "mean_capped_seconds": 61.509375,
              "training_log": "outputs/online/training-v7-broad-r2.json",
              "resume": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
              "datasets": [
                "outputs/online/imitation-initial.npz",
                "outputs/online/imitation-broad.npz",
                "outputs/online/dagger-broad-r1.npz",
                "outputs/online/dagger-broad-r2.npz"
              ]
            },
            {
              "checkpoint": "outputs/online/checkpoint/v7-broad-r3-04000.pt",
              "report": "outputs/online/v7-development-broad-r3-04000.json",
              "width": 1024,
              "round": 3,
              "steps": 4000,
              "successes": 20,
              "trials": 32,
              "mean_capped_seconds": 60.25625,
              "training_log": "outputs/online/training-v7-broad-r3.json",
              "resume": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
              "datasets": [
                "outputs/online/imitation-initial.npz",
                "outputs/online/imitation-broad.npz",
                "outputs/online/dagger-broad-r1.npz",
                "outputs/online/dagger-broad-r2.npz",
                "outputs/online/dagger-broad-r3.npz"
              ]
            },
            {
              "checkpoint": "outputs/online/checkpoint/v7-broad-r3-08000.pt",
              "report": "outputs/online/v7-development-broad-r3-08000.json",
              "width": 1024,
              "round": 3,
              "steps": 8000,
              "successes": 20,
              "trials": 32,
              "mean_capped_seconds": 61.109375,
              "training_log": "outputs/online/training-v7-broad-r3.json",
              "resume": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
              "datasets": [
                "outputs/online/imitation-initial.npz",
                "outputs/online/imitation-broad.npz",
                "outputs/online/dagger-broad-r1.npz",
                "outputs/online/dagger-broad-r2.npz",
                "outputs/online/dagger-broad-r3.npz"
              ]
            }
          ],
          "selected": {
            "checkpoint": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
            "report": "outputs/online/v7-development-broad-r2-08000.json",
            "width": 1024,
            "round": 2,
            "steps": 8000,
            "successes": 22,
            "trials": 32,
            "mean_capped_seconds": 61.509375,
            "training_log": "outputs/online/training-v7-broad-r2.json",
            "resume": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
            "datasets": [
              "outputs/online/imitation-initial.npz",
              "outputs/online/imitation-broad.npz",
              "outputs/online/dagger-broad-r1.npz",
              "outputs/online/dagger-broad-r2.npz"
            ]
          },
          "selected_sha256": "d96d011dfee9f2c6f2e6a511c21c735b258c2ba9f1366e1932edcaa02c37f751"
        },
        "v8": {
          "method": "Highest development success, then lowest capped duration; ties favour fewer updates. Two reward proxies, three checkpoints each; no final tests used.",
          "candidates": [
            {
              "checkpoint": "outputs/online/checkpoint/rl-gpu-100.pt",
              "report": "outputs/online/v8-development-gpu-100.json",
              "updates": 100,
              "environment_steps": 409600,
              "reward_mode": "center",
              "successes": 4,
              "trials": 32,
              "mean_capped_seconds": 106.059375,
              "training_log": "outputs/online/training-rl-gpu.json"
            },
            {
              "checkpoint": "outputs/online/checkpoint/rl-gpu-300.pt",
              "report": "outputs/online/v8-development-gpu-300.json",
              "updates": 300,
              "environment_steps": 1228800,
              "reward_mode": "center",
              "successes": 2,
              "trials": 32,
              "mean_capped_seconds": 112.7375,
              "training_log": "outputs/online/training-rl-gpu.json"
            },
            {
              "checkpoint": "outputs/online/checkpoint/rl-exit-100.pt",
              "report": "outputs/online/v8-development-exit-100.json",
              "updates": 100,
              "environment_steps": 409600,
              "reward_mode": "exit",
              "successes": 3,
              "trials": 32,
              "mean_capped_seconds": 109.325,
              "training_log": "outputs/online/training-rl-exit.json"
            },
            {
              "checkpoint": "outputs/online/checkpoint/rl-gpu-600.pt",
              "report": "outputs/online/v8-development-gpu-600.json",
              "updates": 600,
              "environment_steps": 2457600,
              "reward_mode": "center",
              "successes": 4,
              "trials": 32,
              "mean_capped_seconds": 106.384375,
              "training_log": "outputs/online/training-rl-gpu.json"
            },
            {
              "checkpoint": "outputs/online/checkpoint/rl-exit-300.pt",
              "report": "outputs/online/v8-development-exit-300.json",
              "updates": 300,
              "environment_steps": 1228800,
              "reward_mode": "exit",
              "successes": 1,
              "trials": 32,
              "mean_capped_seconds": 116.353125,
              "training_log": "outputs/online/training-rl-exit.json"
            },
            {
              "checkpoint": "outputs/online/checkpoint/rl-exit-600.pt",
              "report": "outputs/online/v8-development-exit-600.json",
              "updates": 600,
              "environment_steps": 2457600,
              "reward_mode": "exit",
              "successes": 3,
              "trials": 32,
              "mean_capped_seconds": 109.5375,
              "training_log": "outputs/online/training-rl-exit.json"
            }
          ],
          "selected": {
            "checkpoint": "outputs/online/checkpoint/rl-gpu-100.pt",
            "report": "outputs/online/v8-development-gpu-100.json",
            "updates": 100,
            "environment_steps": 409600,
            "reward_mode": "center",
            "successes": 4,
            "trials": 32,
            "mean_capped_seconds": 106.059375,
            "training_log": "outputs/online/training-rl-gpu.json"
          },
          "selected_sha256": "2840b561504867d53321835518821758b406da37e52cdc95c36da1bc0a8f17cc"
        }
      },
      "default_model": "v7"
    },
    "sha256": {
      "dustfly/online_policy.py": "95d4424a53bc9a4b6a9fdc7c7b0dfb241ef3bacf80eaa715ae1fd2a124d87f38",
      "dustfly/online_rl.py": "b09c3f5bdb119b5e126ed01def5f07d55c4670b8254c3bd38ccaf34c97f4228d",
      "dustfly/navigation_online.py": "7f390562d1b1015fa67037dc30d9c77537a4ba32ce2a1ba8667565bcfa37a7ba",
      "dustfly/brain.py": "46000db9fa282d61b82b2db6ca38010eb6d485cd39c973d814c4630a1c217f6c",
      "dustfly/adapted.py": "a4d368bd66e5506d1cfcb6335eb6d49b6bcdbf9cb1f83e32594395c068593fc5",
      "dustfly/plastic.py": "fb19d0729a9ee2643a6ef02705b78e5bf76a503491e836c550c73150a702b63c",
      "dustfly/geometry.py": "2e7e0c8e7da11fdc21da9edfe2edab93f1b9137f5dcd175cb7aea99702cdb589",
      "dustfly/fast_nav.py": "989f58e8a693f2276bde758c5a1f4d9a2101660d2cca27db2642efaabbd7a7b4",
      "dustfly/nav.py": "5a73eb556b9831a284c1441cbdcf9d62af05a3adb92e59ff67612cc71bc05a58",
      "outputs/online/protocol.json": "a95a37ed4ac644e3cb0f7293b2393447f4e2e7a422f6a460142e02aec5049f38",
      "outputs/online/development.json": "b6d55dd901ead18007e0d85fa2f0b9691bd591744e932e4afe4d8960f24e1be2",
      "outputs/online/heldout.json": "9ff459ffeffb369e05b19a2a2c5162828a4f90621e04f89f0a2818e7368f525a",
      "outputs/online/rl-training.json": "91236125406f2be79614405b73bac513b9ce005ab322ab42f5c32b229fa4ed47",
      "outputs/online/showcase.json": "722418b772f07762e31688e2457f52abf3c2eff6246ea6d59e0e22fc971fbf51",
      "outputs/online/selection.json": "708a51d7b1a37d0779940890a23cecefc1783e519ca39bc97e355d056c229feb",
      "outputs/v2/original-parameters.json": "a308538db8f240443c581e919d65804829fc43b1fc7ebf9e7b59a04ce5a51fdb",
      "outputs/v2/checkpoint/motor.pt": "63a8d0dfbfc6e244e93bdcedcf8de84ae3ba070a4408bbd1ae9396e76e4647da",
      "outputs/v2/checkpoint/vision.pt": "1cae21090e5ee94164be7d1e5c5245a7a31ee89ab557cf4eff7c2e7988cf46ca",
      "outputs/v3/checkpoint/internal.pt": "a393b19fb18159067b701b65cc4a5a0aaebd4f9bccd307da0572cb932cfca12a",
      "data/connectome/prepared/manifest.json": "70eae87681808635b00887d94bb84abbf66d8f3d0e54f9f81193c109e4a34e49",
      "data/connectome/prepared/weights.npy": "b08d70bda27a710a98d3e3a3d0087be17ad8dae972872fea07e38e8712eb74ed",
      "data/connectome/prepared/indptr.npy": "2efdd6bcfd538d4a58ff381246075f6d1315417c54c38f39a75ad4386f4b689c",
      "data/connectome/prepared/indices.npy": "917e8d715e4075ea11f785d3568b13c3f4f29030e91241399d232c85f4188851",
      "data/simplified_triangles.npy": "bace33133bc6a0e6c355a2d668bba2ee0f143f60dc4a0eb9cae68e97ab820d3e",
      "de_dust2.nav": "6288d75ed8b2edc792c64a2d016741b281d907f3608e0de32ea4ae120b061df1",
      "scripts/train_online.py": "f18cf16406c048575dbb4f97c0f9a6c9ff7d7e2c88b15820510a6b94b7c991f4",
      "scripts/train_online_rl.py": "e7723eb1e08b457dbfc57cbcfc071a56b42be558dcc2bc888a958958557606ae",
      "scripts/prepare_online.py": "f9343d03901539107d1544a8ba9447492d794c7092604c4c0755970b3ddae13a",
      "scripts/online_expert.py": "5605319c729dfa9bbbbaebcd4780e2fe26536cda8ed8d388a804af0cc628da49",
      "scripts/run_online.py": "989b2de5b4aa42ee788c4d1b5363ee6e6a9bf1fbfec35469edb76007b2b5dc3d",
      "dustfly/online_rl_environment.py": "0de3ef8c9b08a37519a978e30613a6b2828a443758acb9cd06aae5edab1816da",
      "outputs/online/training-evidence.json": "f7e745cb8a11c10dd4a807e2044c00479e90725c9bb08e2d26dabba678c10305",
      "scripts/develop_v6.py": "7f947450b0d20973bc20052cf9902a63d138b836e757c562478a863a93091001",
      "scripts/develop_v7.py": "8bbee4319bf7fb7358d3c0d797b979b6c2063c63a73cbdfa74e128faa43c9266",
      "scripts/select_online_v8.py": "54e5b86733246216ec693b16f00a749a23790f97d55302d749c4b987ba31b45f",
      "scripts/finalize_online.py": "190a60a775aeed9d800667ab8088cdb05aa3c2261e5659a2ab748d38ae075ec5",
      "scripts/freeze_online.py": "a8d244c400aeffe083a120904d12fc3c27ca40f2189c9e81498b5e8f3b5f0220",
      "data/connectome/prepared/optic_indices.npy": "5acf991839fd17f0befc00db3044d29e4b6416fa14bd27a44a77a6edddb4681c",
      "outputs/v4/navigation-teacher.npz": "9c645929d893acf22838e5afc40bba8b9018ea1a830f850997871af0731fe3e3",
      "showcase/public/data/neural-layout.json": "b52a933c764cf3fb205440fbc3336257e06577dc1a5e75aec663973f28175724",
      "showcase/public/data/neuron-interfaces.csv": "6ba22d21a3a259e983f263b60b0659a4f13d6e36998586d71b46a6d382676f52",
      "outputs/online/checkpoint/v6.pt": "3eb1b01a2bcbbe400d86f4b31360434bc61a50c879379f98c6ef82131b2e07f6",
      "outputs/online/checkpoint/v7.pt": "d96d011dfee9f2c6f2e6a511c21c735b258c2ba9f1366e1932edcaa02c37f751",
      "outputs/online/checkpoint/v8.pt": "2840b561504867d53321835518821758b406da37e52cdc95c36da1bc0a8f17cc"
    },
    "final_test_not_used_for_selection": true
  },
  "verification": {
    "passed": true,
    "checks": [
      {
        "check": "frozen files unchanged",
        "count": 44
      },
      {
        "check": "displayed neuron order matches full-graph telemetry indices",
        "neurons": 176,
        "displayed_edges": 240
      },
      {
        "check": "trajectory identity, clocks, physical limits, goal outcome",
        "runs": 396,
        "frames": 346671
      },
      {
        "check": "walkable surface sample",
        "samples_per_run": 7
      },
      {
        "check": "recorded neural/depth encoding and ranges",
        "samples": 20137
      },
      {
        "check": "full graph, runtime information boundary and PFL3-only stopping",
        "models": 3
      }
    ],
    "limits": "Software/numerical and frozen-policy intervention checks, not biological validation or a retrained topology comparison."
  },
  "training": {
    "created_at_utc": "2026-09-13T07:32:26.089770+00:00",
    "models": {
      "v6": {
        "checkpoint": "outputs/online/checkpoint/v6.pt",
        "sha256": "3eb1b01a2bcbbe400d86f4b31360434bc61a50c879379f98c6ef82131b2e07f6",
        "selected_source": "outputs/online/checkpoint/v6-broad-w1024-08000.pt",
        "selected_checkpoint_metadata": {
          "method": "Supervised teacher examples; DAgger when aggregated datasets are supplied",
          "inputs": "Fresh depth image, continuous current XYZ and yaw, continuous goal XYZ",
          "parameters": 1793538,
          "width": 1024,
          "training_examples": 950000,
          "validation_examples": 50000,
          "steps": 8000,
          "batch": 2048,
          "seed": 130941,
          "data": [
            "outputs/online/imitation-initial.npz",
            "outputs/online/imitation-broad.npz"
          ],
          "resume": null,
          "learning_rate": 0.001,
          "inference": "Fresh network forward each 100 ms; no map-output cache, no route planner, no NAV IDs in policy",
          "validation_scope": "Image split is a fitting diagnostic only; navigation validation uses separate development situations."
        },
        "development": {
          "successes": 19,
          "trials": 32,
          "mean_capped_seconds": 67.521875
        },
        "width": 1024,
        "training_examples": 950000,
        "image_validation_examples": 50000,
        "trainable_parameters": 1793538,
        "inference_parameters": 27382366,
        "optimizer_steps": 8000,
        "lineage": [
          {
            "checkpoint": "outputs/online/checkpoint/v6-broad-w1024-08000.pt",
            "sha256": "3eb1b01a2bcbbe400d86f4b31360434bc61a50c879379f98c6ef82131b2e07f6",
            "optimizer_steps": 8000,
            "datasets": [
              "outputs/online/imitation-initial.npz",
              "outputs/online/imitation-broad.npz"
            ],
            "width": 1024,
            "learning_rate": 0.001,
            "dagger_sampling_fraction": 0
          }
        ],
        "lineage_optimizer_steps": 8000
      },
      "v7": {
        "checkpoint": "outputs/online/checkpoint/v7.pt",
        "sha256": "d96d011dfee9f2c6f2e6a511c21c735b258c2ba9f1366e1932edcaa02c37f751",
        "selected_source": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
        "selected_checkpoint_metadata": {
          "method": "Supervised teacher examples; DAgger when aggregated datasets are supplied",
          "inputs": "Fresh depth image, continuous current XYZ and yaw, continuous goal XYZ",
          "parameters": 1793538,
          "width": 1024,
          "dagger_sampling_fraction": 0.5,
          "training_examples": 1020300,
          "validation_examples": 53700,
          "steps": 8000,
          "batch": 2048,
          "seed": 130941,
          "data": [
            "outputs/online/imitation-initial.npz",
            "outputs/online/imitation-broad.npz",
            "outputs/online/dagger-broad-r1.npz",
            "outputs/online/dagger-broad-r2.npz"
          ],
          "resume": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
          "learning_rate": 0.00015,
          "inference": "Fresh network forward each 100 ms; no map-output cache, no route planner, no NAV IDs in policy",
          "validation_scope": "Image split is a fitting diagnostic only; navigation validation uses separate development situations."
        },
        "development": {
          "successes": 22,
          "trials": 32,
          "mean_capped_seconds": 61.509375
        },
        "width": 1024,
        "training_examples": 1020300,
        "image_validation_examples": 53700,
        "trainable_parameters": 1793538,
        "inference_parameters": 27382366,
        "optimizer_steps": 8000,
        "lineage": [
          {
            "checkpoint": "outputs/online/checkpoint/v6-selected-development.pt",
            "sha256": "3eb1b01a2bcbbe400d86f4b31360434bc61a50c879379f98c6ef82131b2e07f6",
            "optimizer_steps": 8000,
            "datasets": [
              "outputs/online/imitation-initial.npz",
              "outputs/online/imitation-broad.npz"
            ],
            "width": 1024,
            "learning_rate": 0.001,
            "dagger_sampling_fraction": 0
          },
          {
            "checkpoint": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
            "sha256": "4a112ef0f67b3ed23ecc62c278451cd1dcf42051d16116db83bfba0d519937b7",
            "optimizer_steps": 4000,
            "datasets": [
              "outputs/online/imitation-initial.npz",
              "outputs/online/imitation-broad.npz",
              "outputs/online/dagger-broad-r1.npz"
            ],
            "width": 1024,
            "learning_rate": 0.0003,
            "dagger_sampling_fraction": 0.5
          },
          {
            "checkpoint": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
            "sha256": "d96d011dfee9f2c6f2e6a511c21c735b258c2ba9f1366e1932edcaa02c37f751",
            "optimizer_steps": 8000,
            "datasets": [
              "outputs/online/imitation-initial.npz",
              "outputs/online/imitation-broad.npz",
              "outputs/online/dagger-broad-r1.npz",
              "outputs/online/dagger-broad-r2.npz"
            ],
            "width": 1024,
            "learning_rate": 0.00015,
            "dagger_sampling_fraction": 0.5
          }
        ],
        "lineage_optimizer_steps": 20000
      },
      "v8": {
        "checkpoint": "outputs/online/checkpoint/v8.pt",
        "sha256": "2840b561504867d53321835518821758b406da37e52cdc95c36da1bc0a8f17cc",
        "selected_source": "outputs/online/checkpoint/rl-gpu-100.pt",
        "selected_checkpoint_metadata": {
          "method": "Joint hierarchical PPO of sensory and motor interfaces from scratch",
          "depth_renderer": "NVIDIA GeForce RTX 5090/PCIe/SSE2",
          "compute_device": "NVIDIA GeForce RTX 5090",
          "seed": 130942,
          "updates": 100,
          "environment_steps": 409600,
          "sensory_parameters": 120577,
          "motor_parameters": 5890,
          "critic_parameters": 120577,
          "exploration_parameters": 3,
          "pretrained_neural_adapters": false,
          "anatomy": "Original fixed full retained MaleCNS graph and calibrated V1 constants; no V2/V3 weights",
          "sensory_inputs": "Fresh depth image, current XYZ and heading, goal XYZ",
          "motor_inputs": "24 PFL3 activities only, scaled by constant 10",
          "joint_log_probability": "log p(sensory cue | image, pose, goal) + log p(motor residual | PFL3 after cue)",
          "brain_gradient": "Brain transition is part of the stochastic control loop; credit assigned with policy gradients, no backpropagation into fixed anatomical weights",
          "reward": "Geodesic progress minus time/blockage penalties plus success bonus and timeout penalty; geometry is reward/physics only",
          "curriculum": "Updates 1\u201325: <=12m/25s; 26\u201350: <=30m/45s; 51\u201375: <=80m/75s; later 75% frozen training goals +25% <=80m, 120s",
          "gamma": 0.995,
          "gae_lambda": 0.95,
          "ppo_clip": 0.2,
          "entropy_coefficient": 0.004,
          "epochs": 4,
          "batch": 32,
          "horizon": 128,
          "learning_rate": 0.0003
        },
        "development": {
          "successes": 4,
          "trials": 32,
          "mean_capped_seconds": 106.059375
        },
        "environment_steps": 409600,
        "ppo_updates": 100,
        "reward_mode": "center",
        "trainable_parameters": 247047,
        "inference_parameters": 25709405
      }
    },
    "studies": {
      "v6": {
        "method": "Highest development success, then lowest capped duration; ties favour smaller width and fewer updates. No final tests used.",
        "candidates": [
          {
            "checkpoint": "outputs/online/checkpoint/v6-broad-w512-08000.pt",
            "report": "outputs/online/v6-development-broad-w512-08000.json",
            "width": 512,
            "steps": 8000,
            "successes": 15,
            "trials": 32,
            "mean_capped_seconds": 77.721875,
            "training_log": "outputs/online/training-v6-broad-w512.json"
          },
          {
            "checkpoint": "outputs/online/checkpoint/v6-broad-w512-16000.pt",
            "report": "outputs/online/v6-development-broad-w512-16000.json",
            "width": 512,
            "steps": 16000,
            "successes": 13,
            "trials": 32,
            "mean_capped_seconds": 83.403125,
            "training_log": "outputs/online/training-v6-broad-w512.json"
          },
          {
            "checkpoint": "outputs/online/checkpoint/v6-broad-w512-24000.pt",
            "report": "outputs/online/v6-development-broad-w512-24000.json",
            "width": 512,
            "steps": 24000,
            "successes": 9,
            "trials": 32,
            "mean_capped_seconds": 94.621875,
            "training_log": "outputs/online/training-v6-broad-w512.json"
          },
          {
            "checkpoint": "outputs/online/checkpoint/v6-broad-w1024-08000.pt",
            "report": "outputs/online/v6-development-broad-w1024-08000.json",
            "width": 1024,
            "steps": 8000,
            "successes": 19,
            "trials": 32,
            "mean_capped_seconds": 67.521875,
            "training_log": "outputs/online/training-v6-broad-w1024.json"
          },
          {
            "checkpoint": "outputs/online/checkpoint/v6-broad-w1024-16000.pt",
            "report": "outputs/online/v6-development-broad-w1024-16000.json",
            "width": 1024,
            "steps": 16000,
            "successes": 15,
            "trials": 32,
            "mean_capped_seconds": 77.215625,
            "training_log": "outputs/online/training-v6-broad-w1024.json"
          },
          {
            "checkpoint": "outputs/online/checkpoint/v6-broad-w1024-24000.pt",
            "report": "outputs/online/v6-development-broad-w1024-24000.json",
            "width": 1024,
            "steps": 24000,
            "successes": 14,
            "trials": 32,
            "mean_capped_seconds": 80.88125,
            "training_log": "outputs/online/training-v6-broad-w1024.json"
          }
        ],
        "selected": {
          "checkpoint": "outputs/online/checkpoint/v6-broad-w1024-08000.pt",
          "report": "outputs/online/v6-development-broad-w1024-08000.json",
          "width": 1024,
          "steps": 8000,
          "successes": 19,
          "trials": 32,
          "mean_capped_seconds": 67.521875,
          "training_log": "outputs/online/training-v6-broad-w1024.json"
        },
        "selected_sha256": "3eb1b01a2bcbbe400d86f4b31360434bc61a50c879379f98c6ef82131b2e07f6"
      },
      "v7": {
        "method": "Highest development success, then lowest capped duration; ties favour earlier round and fewer updates. No final tests used.",
        "v6_start": {
          "checkpoint": "outputs/online/checkpoint/v6-broad-w1024-08000.pt",
          "report": "outputs/online/v6-development-broad-w1024-08000.json",
          "width": 1024,
          "steps": 8000,
          "successes": 19,
          "trials": 32,
          "mean_capped_seconds": 67.521875,
          "training_log": "outputs/online/training-v6-broad-w1024.json"
        },
        "rounds": [
          {
            "iteration": 1,
            "collection_checkpoint": "outputs/online/checkpoint/v6-selected-development.pt",
            "training_entries": 128,
            "training_offset": 0,
            "collection_report": "outputs/online/v7-rl-training-broad-r1.json",
            "selected_for_next_round": {
              "checkpoint": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
              "report": "outputs/online/v7-development-broad-r1-04000.json",
              "width": 1024,
              "round": 1,
              "steps": 4000,
              "successes": 17,
              "trials": 32,
              "mean_capped_seconds": 68.946875,
              "training_log": "outputs/online/training-v7-broad-r1.json",
              "resume": "outputs/online/checkpoint/v6-selected-development.pt",
              "datasets": [
                "outputs/online/imitation-initial.npz",
                "outputs/online/imitation-broad.npz",
                "outputs/online/dagger-broad-r1.npz"
              ]
            }
          },
          {
            "iteration": 2,
            "collection_checkpoint": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
            "training_entries": 128,
            "training_offset": 128,
            "collection_report": "outputs/online/v7-rl-training-broad-r2.json",
            "selected_for_next_round": {
              "checkpoint": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
              "report": "outputs/online/v7-development-broad-r2-08000.json",
              "width": 1024,
              "round": 2,
              "steps": 8000,
              "successes": 22,
              "trials": 32,
              "mean_capped_seconds": 61.509375,
              "training_log": "outputs/online/training-v7-broad-r2.json",
              "resume": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
              "datasets": [
                "outputs/online/imitation-initial.npz",
                "outputs/online/imitation-broad.npz",
                "outputs/online/dagger-broad-r1.npz",
                "outputs/online/dagger-broad-r2.npz"
              ]
            }
          },
          {
            "iteration": 3,
            "collection_checkpoint": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
            "training_entries": 128,
            "training_offset": 256,
            "collection_report": "outputs/online/v7-rl-training-broad-r3.json",
            "selected_for_next_round": {
              "checkpoint": "outputs/online/checkpoint/v7-broad-r3-04000.pt",
              "report": "outputs/online/v7-development-broad-r3-04000.json",
              "width": 1024,
              "round": 3,
              "steps": 4000,
              "successes": 20,
              "trials": 32,
              "mean_capped_seconds": 60.25625,
              "training_log": "outputs/online/training-v7-broad-r3.json",
              "resume": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
              "datasets": [
                "outputs/online/imitation-initial.npz",
                "outputs/online/imitation-broad.npz",
                "outputs/online/dagger-broad-r1.npz",
                "outputs/online/dagger-broad-r2.npz",
                "outputs/online/dagger-broad-r3.npz"
              ]
            }
          }
        ],
        "candidates": [
          {
            "checkpoint": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
            "report": "outputs/online/v7-development-broad-r1-04000.json",
            "width": 1024,
            "round": 1,
            "steps": 4000,
            "successes": 17,
            "trials": 32,
            "mean_capped_seconds": 68.946875,
            "training_log": "outputs/online/training-v7-broad-r1.json",
            "resume": "outputs/online/checkpoint/v6-selected-development.pt",
            "datasets": [
              "outputs/online/imitation-initial.npz",
              "outputs/online/imitation-broad.npz",
              "outputs/online/dagger-broad-r1.npz"
            ]
          },
          {
            "checkpoint": "outputs/online/checkpoint/v7-broad-r1-08000.pt",
            "report": "outputs/online/v7-development-broad-r1-08000.json",
            "width": 1024,
            "round": 1,
            "steps": 8000,
            "successes": 14,
            "trials": 32,
            "mean_capped_seconds": 76.2,
            "training_log": "outputs/online/training-v7-broad-r1.json",
            "resume": "outputs/online/checkpoint/v6-selected-development.pt",
            "datasets": [
              "outputs/online/imitation-initial.npz",
              "outputs/online/imitation-broad.npz",
              "outputs/online/dagger-broad-r1.npz"
            ]
          },
          {
            "checkpoint": "outputs/online/checkpoint/v7-broad-r2-04000.pt",
            "report": "outputs/online/v7-development-broad-r2-04000.json",
            "width": 1024,
            "round": 2,
            "steps": 4000,
            "successes": 20,
            "trials": 32,
            "mean_capped_seconds": 61.696875,
            "training_log": "outputs/online/training-v7-broad-r2.json",
            "resume": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
            "datasets": [
              "outputs/online/imitation-initial.npz",
              "outputs/online/imitation-broad.npz",
              "outputs/online/dagger-broad-r1.npz",
              "outputs/online/dagger-broad-r2.npz"
            ]
          },
          {
            "checkpoint": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
            "report": "outputs/online/v7-development-broad-r2-08000.json",
            "width": 1024,
            "round": 2,
            "steps": 8000,
            "successes": 22,
            "trials": 32,
            "mean_capped_seconds": 61.509375,
            "training_log": "outputs/online/training-v7-broad-r2.json",
            "resume": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
            "datasets": [
              "outputs/online/imitation-initial.npz",
              "outputs/online/imitation-broad.npz",
              "outputs/online/dagger-broad-r1.npz",
              "outputs/online/dagger-broad-r2.npz"
            ]
          },
          {
            "checkpoint": "outputs/online/checkpoint/v7-broad-r3-04000.pt",
            "report": "outputs/online/v7-development-broad-r3-04000.json",
            "width": 1024,
            "round": 3,
            "steps": 4000,
            "successes": 20,
            "trials": 32,
            "mean_capped_seconds": 60.25625,
            "training_log": "outputs/online/training-v7-broad-r3.json",
            "resume": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
            "datasets": [
              "outputs/online/imitation-initial.npz",
              "outputs/online/imitation-broad.npz",
              "outputs/online/dagger-broad-r1.npz",
              "outputs/online/dagger-broad-r2.npz",
              "outputs/online/dagger-broad-r3.npz"
            ]
          },
          {
            "checkpoint": "outputs/online/checkpoint/v7-broad-r3-08000.pt",
            "report": "outputs/online/v7-development-broad-r3-08000.json",
            "width": 1024,
            "round": 3,
            "steps": 8000,
            "successes": 20,
            "trials": 32,
            "mean_capped_seconds": 61.109375,
            "training_log": "outputs/online/training-v7-broad-r3.json",
            "resume": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
            "datasets": [
              "outputs/online/imitation-initial.npz",
              "outputs/online/imitation-broad.npz",
              "outputs/online/dagger-broad-r1.npz",
              "outputs/online/dagger-broad-r2.npz",
              "outputs/online/dagger-broad-r3.npz"
            ]
          }
        ],
        "selected": {
          "checkpoint": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
          "report": "outputs/online/v7-development-broad-r2-08000.json",
          "width": 1024,
          "round": 2,
          "steps": 8000,
          "successes": 22,
          "trials": 32,
          "mean_capped_seconds": 61.509375,
          "training_log": "outputs/online/training-v7-broad-r2.json",
          "resume": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
          "datasets": [
            "outputs/online/imitation-initial.npz",
            "outputs/online/imitation-broad.npz",
            "outputs/online/dagger-broad-r1.npz",
            "outputs/online/dagger-broad-r2.npz"
          ]
        },
        "selected_sha256": "d96d011dfee9f2c6f2e6a511c21c735b258c2ba9f1366e1932edcaa02c37f751"
      },
      "v8": {
        "method": "Highest development success, then lowest capped duration; ties favour fewer updates. Two reward proxies, three checkpoints each; no final tests used.",
        "candidates": [
          {
            "checkpoint": "outputs/online/checkpoint/rl-gpu-100.pt",
            "report": "outputs/online/v8-development-gpu-100.json",
            "updates": 100,
            "environment_steps": 409600,
            "reward_mode": "center",
            "successes": 4,
            "trials": 32,
            "mean_capped_seconds": 106.059375,
            "training_log": "outputs/online/training-rl-gpu.json"
          },
          {
            "checkpoint": "outputs/online/checkpoint/rl-gpu-300.pt",
            "report": "outputs/online/v8-development-gpu-300.json",
            "updates": 300,
            "environment_steps": 1228800,
            "reward_mode": "center",
            "successes": 2,
            "trials": 32,
            "mean_capped_seconds": 112.7375,
            "training_log": "outputs/online/training-rl-gpu.json"
          },
          {
            "checkpoint": "outputs/online/checkpoint/rl-exit-100.pt",
            "report": "outputs/online/v8-development-exit-100.json",
            "updates": 100,
            "environment_steps": 409600,
            "reward_mode": "exit",
            "successes": 3,
            "trials": 32,
            "mean_capped_seconds": 109.325,
            "training_log": "outputs/online/training-rl-exit.json"
          },
          {
            "checkpoint": "outputs/online/checkpoint/rl-gpu-600.pt",
            "report": "outputs/online/v8-development-gpu-600.json",
            "updates": 600,
            "environment_steps": 2457600,
            "reward_mode": "center",
            "successes": 4,
            "trials": 32,
            "mean_capped_seconds": 106.384375,
            "training_log": "outputs/online/training-rl-gpu.json"
          },
          {
            "checkpoint": "outputs/online/checkpoint/rl-exit-300.pt",
            "report": "outputs/online/v8-development-exit-300.json",
            "updates": 300,
            "environment_steps": 1228800,
            "reward_mode": "exit",
            "successes": 1,
            "trials": 32,
            "mean_capped_seconds": 116.353125,
            "training_log": "outputs/online/training-rl-exit.json"
          },
          {
            "checkpoint": "outputs/online/checkpoint/rl-exit-600.pt",
            "report": "outputs/online/v8-development-exit-600.json",
            "updates": 600,
            "environment_steps": 2457600,
            "reward_mode": "exit",
            "successes": 3,
            "trials": 32,
            "mean_capped_seconds": 109.5375,
            "training_log": "outputs/online/training-rl-exit.json"
          }
        ],
        "selected": {
          "checkpoint": "outputs/online/checkpoint/rl-gpu-100.pt",
          "report": "outputs/online/v8-development-gpu-100.json",
          "updates": 100,
          "environment_steps": 409600,
          "reward_mode": "center",
          "successes": 4,
          "trials": 32,
          "mean_capped_seconds": 106.059375,
          "training_log": "outputs/online/training-rl-gpu.json"
        },
        "selected_sha256": "2840b561504867d53321835518821758b406da37e52cdc95c36da1bc0a8f17cc"
      }
    },
    "learning_curves": {
      "outputs/online/training-rl-exit.json": {
        "sha256": "f6d90250541b35669932da7be2b1540e0757bc8f762e0d0550312bdfaa9f9c77",
        "metadata": {
          "method": "Joint hierarchical PPO of sensory and motor interfaces from scratch",
          "depth_renderer": "NVIDIA GeForce RTX 5090/PCIe/SSE2",
          "compute_device": "NVIDIA GeForce RTX 5090",
          "seed": 130942,
          "updates": 600,
          "environment_steps": 2457600,
          "sensory_parameters": 120577,
          "motor_parameters": 5890,
          "critic_parameters": 120577,
          "exploration_parameters": 3,
          "pretrained_neural_adapters": false,
          "anatomy": "Original fixed full retained MaleCNS graph and calibrated V1 constants; no V2/V3 weights",
          "sensory_inputs": "Fresh depth image, current XYZ and heading, goal XYZ",
          "motor_inputs": "24 PFL3 activities only, scaled by constant 10",
          "joint_log_probability": "log p(sensory cue | image, pose, goal) + log p(motor residual | PFL3 after cue)",
          "brain_gradient": "Brain transition is part of the stochastic control loop; credit assigned with policy gradients, no backpropagation into fixed anatomical weights",
          "reward_mode": "exit",
          "reward": "Geodesic progress minus time/blockage penalties plus success bonus and timeout penalty; geometry is reward/physics only",
          "curriculum": "Updates 1\u201325: <=12m/25s; 26\u201350: <=30m/45s; 51\u201375: <=80m/75s; later 75% frozen training goals +25% <=80m, 120s",
          "gamma": 0.995,
          "gae_lambda": 0.95,
          "ppo_clip": 0.2,
          "entropy_coefficient": 0.004,
          "epochs": 4,
          "batch": 32,
          "horizon": 128,
          "learning_rate": 0.0003
        },
        "trace": [
          {
            "update": 1,
            "environment_steps": 4096,
            "curriculum": 0,
            "completed_episodes": 50,
            "new_episodes": 50,
            "new_successes": 50,
            "mean_step_reward": 0.09744856506586075,
            "loss": 15.321024358272552,
            "approx_kl": 0.0017600053924979875,
            "wall_seconds": 10.702140899986262
          },
          {
            "update": 2,
            "environment_steps": 8192,
            "curriculum": 0,
            "completed_episodes": 86,
            "new_episodes": 36,
            "new_successes": 33,
            "mean_step_reward": -0.03410186991095543,
            "loss": 11.683910250663757,
            "approx_kl": 0.0005928421905991854,
            "wall_seconds": 17.620908799988683
          },
          {
            "update": 3,
            "environment_steps": 12288,
            "curriculum": 0,
            "completed_episodes": 132,
            "new_episodes": 46,
            "new_successes": 36,
            "mean_step_reward": -0.002735644578933716,
            "loss": 11.611922264099121,
            "approx_kl": 0.0017338248308078619,
            "wall_seconds": 24.10417310000048
          },
          {
            "update": 4,
            "environment_steps": 16384,
            "curriculum": 0,
            "completed_episodes": 181,
            "new_episodes": 49,
            "new_successes": 44,
            "mean_step_reward": 0.06407812982797623,
            "loss": 11.885345339775085,
            "approx_kl": 0.0014706913716509007,
            "wall_seconds": 30.579708199977176
          },
          {
            "update": 5,
            "environment_steps": 20480,
            "curriculum": 0,
            "completed_episodes": 221,
            "new_episodes": 40,
            "new_successes": 34,
            "mean_step_reward": -0.03088703379034996,
            "loss": 10.998957216739655,
            "approx_kl": 0.0020065672761120368,
            "wall_seconds": 36.96997229999397
          },
          {
            "update": 6,
            "environment_steps": 24576,
            "curriculum": 0,
            "completed_episodes": 268,
            "new_episodes": 47,
            "new_successes": 39,
            "mean_step_reward": 0.02071475051343441,
            "loss": 10.444232761859894,
            "approx_kl": 0.0018193202904512873,
            "wall_seconds": 43.42837119998876
          },
          {
            "update": 7,
            "environment_steps": 28672,
            "curriculum": 0,
            "completed_episodes": 318,
            "new_episodes": 50,
            "new_successes": 46,
            "mean_step_reward": 0.06414201110601425,
            "loss": 11.684995353221893,
            "approx_kl": 0.0028825220451835776,
            "wall_seconds": 49.876373999984935
          },
          {
            "update": 8,
            "environment_steps": 32768,
            "curriculum": 0,
            "completed_episodes": 369,
            "new_episodes": 51,
            "new_successes": 45,
            "mean_step_reward": 0.08306959271430969,
            "loss": 10.105521261692047,
            "approx_kl": 0.0011831748961412814,
            "wall_seconds": 56.3376512999821
          },
          {
            "update": 9,
            "environment_steps": 36864,
            "curriculum": 0,
            "completed_episodes": 431,
            "new_episodes": 62,
            "new_successes": 57,
            "mean_step_reward": 0.14949709177017212,
            "loss": 11.619970560073853,
            "approx_kl": 0.0020576234855980147,
            "wall_seconds": 62.743081699998584
          },
          {
            "update": 10,
            "environment_steps": 40960,
            "curriculum": 0,
            "completed_episodes": 475,
            "new_episodes": 44,
            "new_successes": 38,
            "mean_step_reward": 0.02652323804795742,
            "loss": 8.448331207036972,
            "approx_kl": 0.001965693880265462,
            "wall_seconds": 69.18119730000035
          },
          {
            "update": 11,
            "environment_steps": 45056,
            "curriculum": 0,
            "completed_episodes": 524,
            "new_episodes": 49,
            "new_successes": 46,
            "mean_step_reward": 0.0845499113202095,
            "loss": 9.96539056301117,
            "approx_kl": 0.0006703132748953067,
            "wall_seconds": 75.71669659999316
          },
          {
            "update": 12,
            "environment_steps": 49152,
            "curriculum": 0,
            "completed_episodes": 579,
            "new_episodes": 55,
            "new_successes": 48,
            "mean_step_reward": 0.08191979676485062,
            "loss": 9.543879449367523,
            "approx_kl": 0.0011513199242472183,
            "wall_seconds": 82.11188389998279
          },
          {
            "update": 13,
            "environment_steps": 53248,
            "curriculum": 0,
            "completed_episodes": 632,
            "new_episodes": 53,
            "new_successes": 47,
            "mean_step_reward": 0.09866801649332047,
            "loss": 11.097898423671722,
            "approx_kl": 0.0015663144768041093,
            "wall_seconds": 88.60888699998031
          },
          {
            "update": 14,
            "environment_steps": 57344,
            "curriculum": 0,
            "completed_episodes": 671,
            "new_episodes": 39,
            "new_successes": 33,
            "mean_step_reward": -0.013950321823358536,
            "loss": 8.790927231311798,
            "approx_kl": 0.0010367714767198777,
            "wall_seconds": 95.10361389999161
          },
          {
            "update": 15,
            "environment_steps": 61440,
            "curriculum": 0,
            "completed_episodes": 721,
            "new_episodes": 50,
            "new_successes": 42,
            "mean_step_reward": 0.02483554370701313,
            "loss": 13.234204947948456,
            "approx_kl": 0.0011843118500109995,
            "wall_seconds": 101.03724429997965
          },
          {
            "update": 16,
            "environment_steps": 65536,
            "curriculum": 0,
            "completed_episodes": 759,
            "new_episodes": 38,
            "new_successes": 34,
            "mean_step_reward": -0.0017154961824417114,
            "loss": 11.1582692861557,
            "approx_kl": 0.00027093366588815115,
            "wall_seconds": 106.89696049998747
          },
          {
            "update": 17,
            "environment_steps": 69632,
            "curriculum": 0,
            "completed_episodes": 810,
            "new_episodes": 51,
            "new_successes": 45,
            "mean_step_reward": 0.08233332633972168,
            "loss": 11.074466526508331,
            "approx_kl": 0.0023154337613959797,
            "wall_seconds": 112.79563049998251
          },
          {
            "update": 18,
            "environment_steps": 73728,
            "curriculum": 0,
            "completed_episodes": 862,
            "new_episodes": 52,
            "new_successes": 47,
            "mean_step_reward": 0.11290116608142853,
            "loss": 9.721475332975388,
            "approx_kl": 0.0010611624038574519,
            "wall_seconds": 118.66810339997755
          },
          {
            "update": 19,
            "environment_steps": 77824,
            "curriculum": 0,
            "completed_episodes": 915,
            "new_episodes": 53,
            "new_successes": 48,
            "mean_step_reward": 0.10426685214042664,
            "loss": 10.69073623418808,
            "approx_kl": 0.001607793015864445,
            "wall_seconds": 124.65089940000325
          },
          {
            "update": 20,
            "environment_steps": 81920,
            "curriculum": 0,
            "completed_episodes": 965,
            "new_episodes": 50,
            "new_successes": 41,
            "mean_step_reward": 0.05227725952863693,
            "loss": 8.412779808044434,
            "approx_kl": 0.0008651698117319029,
            "wall_seconds": 131.18604430000414
          },
          {
            "update": 21,
            "environment_steps": 86016,
            "curriculum": 0,
            "completed_episodes": 1013,
            "new_episodes": 48,
            "new_successes": 44,
            "mean_step_reward": 0.04035334289073944,
            "loss": 8.914425194263458,
            "approx_kl": 0.00035227727312303614,
            "wall_seconds": 139.19313050000346
          },
          {
            "update": 22,
            "environment_steps": 90112,
            "curriculum": 0,
            "completed_episodes": 1068,
            "new_episodes": 55,
            "new_successes": 45,
            "mean_step_reward": 0.062416110187768936,
            "loss": 10.813582360744476,
            "approx_kl": 0.0011953127232118277,
            "wall_seconds": 147.2240993999876
          },
          {
            "update": 23,
            "environment_steps": 94208,
            "curriculum": 0,
            "completed_episodes": 1104,
            "new_episodes": 36,
            "new_successes": 34,
            "mean_step_reward": -0.03408898413181305,
            "loss": 9.124492526054382,
            "approx_kl": 0.0007107159472070634,
            "wall_seconds": 155.23181909997948
          },
          {
            "update": 24,
            "environment_steps": 98304,
            "curriculum": 0,
            "completed_episodes": 1161,
            "new_episodes": 57,
            "new_successes": 50,
            "mean_step_reward": 0.12867864966392517,
            "loss": 12.792930960655212,
            "approx_kl": 0.0023101894512365106,
            "wall_seconds": 163.34215589999803
          },
          {
            "update": 25,
            "environment_steps": 102400,
            "curriculum": 0,
            "completed_episodes": 1217,
            "new_episodes": 56,
            "new_successes": 52,
            "mean_step_reward": 0.11794524639844894,
            "loss": 8.462897330522537,
            "approx_kl": 0.0016281061834888533,
            "wall_seconds": 171.35820879999665
          },
          {
            "update": 26,
            "environment_steps": 106496,
            "curriculum": 1,
            "completed_episodes": 1246,
            "new_episodes": 29,
            "new_successes": 29,
            "mean_step_reward": 0.053619109094142914,
            "loss": 7.897998481988907,
            "approx_kl": 0.002558109314122703,
            "wall_seconds": 179.31986639997922
          },
          {
            "update": 27,
            "environment_steps": 110592,
            "curriculum": 1,
            "completed_episodes": 1270,
            "new_episodes": 24,
            "new_successes": 20,
            "mean_step_reward": -0.11513395607471466,
            "loss": 6.4670113027095795,
            "approx_kl": 0.0004984333154425258,
            "wall_seconds": 186.34192139998777
          },
          {
            "update": 28,
            "environment_steps": 114688,
            "curriculum": 1,
            "completed_episodes": 1282,
            "new_episodes": 12,
            "new_successes": 10,
            "mean_step_reward": -0.158274844288826,
            "loss": 6.100626319646835,
            "approx_kl": 0.0009286703116231365,
            "wall_seconds": 194.7906608999765
          },
          {
            "update": 29,
            "environment_steps": 118784,
            "curriculum": 1,
            "completed_episodes": 1294,
            "new_episodes": 12,
            "new_successes": 9,
            "mean_step_reward": -0.21752774715423584,
            "loss": 6.465431451797485,
            "approx_kl": 0.0005546890333789634,
            "wall_seconds": 203.98930069999187
          },
          {
            "update": 30,
            "environment_steps": 122880,
            "curriculum": 1,
            "completed_episodes": 1314,
            "new_episodes": 20,
            "new_successes": 9,
            "mean_step_reward": -0.1979144662618637,
            "loss": 7.215070128440857,
            "approx_kl": 0.00027394086828280706,
            "wall_seconds": 213.25705099999323
          },
          {
            "update": 31,
            "environment_steps": 126976,
            "curriculum": 1,
            "completed_episodes": 1334,
            "new_episodes": 20,
            "new_successes": 15,
            "mean_step_reward": -0.15761588513851166,
            "loss": 9.977306127548218,
            "approx_kl": 0.001075041127478471,
            "wall_seconds": 222.58971549998387
          },
          {
            "update": 32,
            "environment_steps": 131072,
            "curriculum": 1,
            "completed_episodes": 1354,
            "new_episodes": 20,
            "new_successes": 13,
            "mean_step_reward": -0.14243653416633606,
            "loss": 9.24878466129303,
            "approx_kl": 0.003279450456830091,
            "wall_seconds": 231.944250200002
          },
          {
            "update": 33,
            "environment_steps": 135168,
            "curriculum": 1,
            "completed_episodes": 1374,
            "new_episodes": 20,
            "new_successes": 15,
            "mean_step_reward": -0.12285754084587097,
            "loss": 9.13039842247963,
            "approx_kl": 0.001237032922290382,
            "wall_seconds": 241.29458819999127
          },
          {
            "update": 34,
            "environment_steps": 139264,
            "curriculum": 1,
            "completed_episodes": 1392,
            "new_episodes": 18,
            "new_successes": 12,
            "mean_step_reward": -0.17100706696510315,
            "loss": 8.887889683246613,
            "approx_kl": 0.0012658608302444918,
            "wall_seconds": 250.6477683999983
          },
          {
            "update": 35,
            "environment_steps": 143360,
            "curriculum": 1,
            "completed_episodes": 1410,
            "new_episodes": 18,
            "new_successes": 13,
            "mean_step_reward": -0.16752150654792786,
            "loss": 10.275283843278885,
            "approx_kl": 0.002432633920761873,
            "wall_seconds": 258.65737199998694
          },
          {
            "update": 36,
            "environment_steps": 147456,
            "curriculum": 1,
            "completed_episodes": 1437,
            "new_episodes": 27,
            "new_successes": 22,
            "mean_step_reward": -0.03366948664188385,
            "loss": 12.457463502883911,
            "approx_kl": 0.001424469897756353,
            "wall_seconds": 267.0391845999984
          },
          {
            "update": 37,
            "environment_steps": 151552,
            "curriculum": 1,
            "completed_episodes": 1461,
            "new_episodes": 24,
            "new_successes": 18,
            "mean_step_reward": -0.05540797859430313,
            "loss": 14.410191655158997,
            "approx_kl": 0.0024536803121009143,
            "wall_seconds": 276.3103885999881
          },
          {
            "update": 38,
            "environment_steps": 155648,
            "curriculum": 1,
            "completed_episodes": 1488,
            "new_episodes": 27,
            "new_successes": 22,
            "mean_step_reward": -0.05784552916884422,
            "loss": 12.168053686618805,
            "approx_kl": 0.0016647077882225858,
            "wall_seconds": 285.6089418999909
          },
          {
            "update": 39,
            "environment_steps": 159744,
            "curriculum": 1,
            "completed_episodes": 1510,
            "new_episodes": 22,
            "new_successes": 18,
            "mean_step_reward": -0.11699166893959045,
            "loss": 8.8990119099617,
            "approx_kl": 0.002433086321616429,
            "wall_seconds": 294.88381859997753
          },
          {
            "update": 40,
            "environment_steps": 163840,
            "curriculum": 1,
            "completed_episodes": 1530,
            "new_episodes": 20,
            "new_successes": 17,
            "mean_step_reward": -0.11734539270401001,
            "loss": 10.174261093139648,
            "approx_kl": 0.0026682262432586867,
            "wall_seconds": 304.26711689998046
          },
          {
            "update": 41,
            "environment_steps": 167936,
            "curriculum": 1,
            "completed_episodes": 1554,
            "new_episodes": 24,
            "new_successes": 17,
            "mean_step_reward": -0.12763667106628418,
            "loss": 8.870830684900284,
            "approx_kl": 0.0006880889013700653,
            "wall_seconds": 313.71995709999464
          },
          {
            "update": 42,
            "environment_steps": 172032,
            "curriculum": 1,
            "completed_episodes": 1574,
            "new_episodes": 20,
            "new_successes": 16,
            "mean_step_reward": -0.13258346915245056,
            "loss": 9.179009646177292,
            "approx_kl": 0.0015082724730746122,
            "wall_seconds": 323.0467411999998
          },
          {
            "update": 43,
            "environment_steps": 176128,
            "curriculum": 1,
            "completed_episodes": 1586,
            "new_episodes": 12,
            "new_successes": 8,
            "mean_step_reward": -0.1948641836643219,
            "loss": 7.952219575643539,
            "approx_kl": 0.0018868306688091252,
            "wall_seconds": 331.11973869998474
          },
          {
            "update": 44,
            "environment_steps": 180224,
            "curriculum": 1,
            "completed_episodes": 1612,
            "new_episodes": 26,
            "new_successes": 19,
            "mean_step_reward": -0.13182908296585083,
            "loss": 12.323527097702026,
            "approx_kl": 0.0014360199038492283,
            "wall_seconds": 339.4199968999892
          },
          {
            "update": 45,
            "environment_steps": 184320,
            "curriculum": 1,
            "completed_episodes": 1626,
            "new_episodes": 14,
            "new_successes": 8,
            "mean_step_reward": -0.1965206265449524,
            "loss": 8.110023766756058,
            "approx_kl": 0.0014016839031683048,
            "wall_seconds": 348.6463521999831
          },
          {
            "update": 46,
            "environment_steps": 188416,
            "curriculum": 1,
            "completed_episodes": 1647,
            "new_episodes": 21,
            "new_successes": 14,
            "mean_step_reward": -0.14763565361499786,
            "loss": 10.412666261196136,
            "approx_kl": 0.0025586946358089335,
            "wall_seconds": 357.9280934999988
          },
          {
            "update": 47,
            "environment_steps": 192512,
            "curriculum": 1,
            "completed_episodes": 1673,
            "new_episodes": 26,
            "new_successes": 19,
            "mean_step_reward": -0.1315411925315857,
            "loss": 10.681669652462006,
            "approx_kl": 0.0029219051994004985,
            "wall_seconds": 367.2083830999909
          },
          {
            "update": 48,
            "environment_steps": 196608,
            "curriculum": 1,
            "completed_episodes": 1692,
            "new_episodes": 19,
            "new_successes": 12,
            "mean_step_reward": -0.15444216132164001,
            "loss": 8.520728766918182,
            "approx_kl": 0.0035948283493780764,
            "wall_seconds": 376.6310792999866
          },
          {
            "update": 49,
            "environment_steps": 200704,
            "curriculum": 1,
            "completed_episodes": 1717,
            "new_episodes": 25,
            "new_successes": 20,
            "mean_step_reward": -0.0911935418844223,
            "loss": 14.31127393245697,
            "approx_kl": 0.0026098157177329995,
            "wall_seconds": 386.0598202999972
          },
          {
            "update": 50,
            "environment_steps": 204800,
            "curriculum": 1,
            "completed_episodes": 1738,
            "new_episodes": 21,
            "new_successes": 16,
            "mean_step_reward": -0.15101048350334167,
            "loss": 10.39123910665512,
            "approx_kl": 0.0026620415974321077,
            "wall_seconds": 395.40420309998444
          },
          {
            "update": 51,
            "environment_steps": 208896,
            "curriculum": 2,
            "completed_episodes": 1748,
            "new_episodes": 10,
            "new_successes": 10,
            "mean_step_reward": -0.1774429976940155,
            "loss": 9.401927590370178,
            "approx_kl": 0.002728232222580118,
            "wall_seconds": 403.62707409998984
          },
          {
            "update": 52,
            "environment_steps": 212992,
            "curriculum": 2,
            "completed_episodes": 1749,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.2828681468963623,
            "loss": 2.741011954843998,
            "approx_kl": 0.0016893464162421878,
            "wall_seconds": 410.35719739997876
          },
          {
            "update": 53,
            "environment_steps": 217088,
            "curriculum": 2,
            "completed_episodes": 1760,
            "new_episodes": 11,
            "new_successes": 4,
            "mean_step_reward": -0.26663827896118164,
            "loss": 4.429885447025299,
            "approx_kl": 0.0021395399126049597,
            "wall_seconds": 419.564799099986
          },
          {
            "update": 54,
            "environment_steps": 221184,
            "curriculum": 2,
            "completed_episodes": 1767,
            "new_episodes": 7,
            "new_successes": 5,
            "mean_step_reward": -0.25266969203948975,
            "loss": 4.677179962396622,
            "approx_kl": 0.001966289984920877,
            "wall_seconds": 428.813616200001
          },
          {
            "update": 55,
            "environment_steps": 225280,
            "curriculum": 2,
            "completed_episodes": 1778,
            "new_episodes": 11,
            "new_successes": 3,
            "mean_step_reward": -0.27281448245048523,
            "loss": 5.621713161468506,
            "approx_kl": 0.0027534856108104577,
            "wall_seconds": 438.07439900000463
          },
          {
            "update": 56,
            "environment_steps": 229376,
            "curriculum": 2,
            "completed_episodes": 1788,
            "new_episodes": 10,
            "new_successes": 3,
            "mean_step_reward": -0.25602155923843384,
            "loss": 6.505840927362442,
            "approx_kl": 0.003116986017630552,
            "wall_seconds": 447.3866607999953
          },
          {
            "update": 57,
            "environment_steps": 233472,
            "curriculum": 2,
            "completed_episodes": 1794,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2703549861907959,
            "loss": 5.449892342090607,
            "approx_kl": 0.002055686120002065,
            "wall_seconds": 456.6550668999844
          },
          {
            "update": 58,
            "environment_steps": 237568,
            "curriculum": 2,
            "completed_episodes": 1796,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2920680642127991,
            "loss": 2.613503083586693,
            "approx_kl": 0.0026699051559262443,
            "wall_seconds": 463.70328169999993
          },
          {
            "update": 59,
            "environment_steps": 241664,
            "curriculum": 2,
            "completed_episodes": 1802,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.31699028611183167,
            "loss": 3.4101777374744415,
            "approx_kl": 0.003626717783845379,
            "wall_seconds": 472.75509939997573
          },
          {
            "update": 60,
            "environment_steps": 245760,
            "curriculum": 2,
            "completed_episodes": 1812,
            "new_episodes": 10,
            "new_successes": 3,
            "mean_step_reward": -0.26962730288505554,
            "loss": 6.001029968261719,
            "approx_kl": 0.002850277744073537,
            "wall_seconds": 481.9530390999862
          },
          {
            "update": 61,
            "environment_steps": 249856,
            "curriculum": 2,
            "completed_episodes": 1824,
            "new_episodes": 12,
            "new_successes": 5,
            "mean_step_reward": -0.2575337290763855,
            "loss": 10.461695849895477,
            "approx_kl": 0.004857376241488964,
            "wall_seconds": 491.20831749998615
          },
          {
            "update": 62,
            "environment_steps": 253952,
            "curriculum": 2,
            "completed_episodes": 1834,
            "new_episodes": 10,
            "new_successes": 5,
            "mean_step_reward": -0.17304649949073792,
            "loss": 11.153573393821716,
            "approx_kl": 0.0030008887479198165,
            "wall_seconds": 500.4886340999801
          },
          {
            "update": 63,
            "environment_steps": 258048,
            "curriculum": 2,
            "completed_episodes": 1842,
            "new_episodes": 8,
            "new_successes": 4,
            "mean_step_reward": -0.19019317626953125,
            "loss": 6.873918384313583,
            "approx_kl": 0.002612858907014015,
            "wall_seconds": 509.8128701999958
          },
          {
            "update": 64,
            "environment_steps": 262144,
            "curriculum": 2,
            "completed_episodes": 1850,
            "new_episodes": 8,
            "new_successes": 6,
            "mean_step_reward": -0.17907148599624634,
            "loss": 9.19403749704361,
            "approx_kl": 0.0036538051990646636,
            "wall_seconds": 519.1568073999952
          },
          {
            "update": 65,
            "environment_steps": 266240,
            "curriculum": 2,
            "completed_episodes": 1857,
            "new_episodes": 7,
            "new_successes": 4,
            "mean_step_reward": -0.22838172316551208,
            "loss": 8.842164903879166,
            "approx_kl": 0.0031000938615761697,
            "wall_seconds": 528.4946221999999
          },
          {
            "update": 66,
            "environment_steps": 270336,
            "curriculum": 2,
            "completed_episodes": 1864,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.28036606311798096,
            "loss": 4.4891724437475204,
            "approx_kl": 0.0033845250036392827,
            "wall_seconds": 535.5970112999785
          },
          {
            "update": 67,
            "environment_steps": 274432,
            "curriculum": 2,
            "completed_episodes": 1876,
            "new_episodes": 12,
            "new_successes": 4,
            "mean_step_reward": -0.2222175896167755,
            "loss": 8.01357352733612,
            "approx_kl": 0.004983687449566787,
            "wall_seconds": 544.7336122999841
          },
          {
            "update": 68,
            "environment_steps": 278528,
            "curriculum": 2,
            "completed_episodes": 1881,
            "new_episodes": 5,
            "new_successes": 3,
            "mean_step_reward": -0.23908717930316925,
            "loss": 4.82700152695179,
            "approx_kl": 0.0032321666039933916,
            "wall_seconds": 553.9129071999923
          },
          {
            "update": 69,
            "environment_steps": 282624,
            "curriculum": 2,
            "completed_episodes": 1887,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.2956687808036804,
            "loss": 7.747179418802261,
            "approx_kl": 0.004046250198371126,
            "wall_seconds": 563.1794632999809
          },
          {
            "update": 70,
            "environment_steps": 286720,
            "curriculum": 2,
            "completed_episodes": 1891,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.30676883459091187,
            "loss": 3.6478024274110794,
            "approx_kl": 0.0016244294074567733,
            "wall_seconds": 572.4441066999862
          },
          {
            "update": 71,
            "environment_steps": 290816,
            "curriculum": 2,
            "completed_episodes": 1900,
            "new_episodes": 9,
            "new_successes": 1,
            "mean_step_reward": -0.29252153635025024,
            "loss": 8.483678221702576,
            "approx_kl": 0.004004464293757337,
            "wall_seconds": 581.7617510999844
          },
          {
            "update": 72,
            "environment_steps": 294912,
            "curriculum": 2,
            "completed_episodes": 1903,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2684951424598694,
            "loss": 3.0752753615379333,
            "approx_kl": 0.00246971150409081,
            "wall_seconds": 591.0620352999831
          },
          {
            "update": 73,
            "environment_steps": 299008,
            "curriculum": 2,
            "completed_episodes": 1914,
            "new_episodes": 11,
            "new_successes": 2,
            "mean_step_reward": -0.31039461493492126,
            "loss": 6.328065484762192,
            "approx_kl": 0.004133557287786971,
            "wall_seconds": 600.3891569999978
          },
          {
            "update": 74,
            "environment_steps": 303104,
            "curriculum": 2,
            "completed_episodes": 1919,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.25808048248291016,
            "loss": 5.442036330699921,
            "approx_kl": 0.0025519804203213425,
            "wall_seconds": 607.5069620999857
          },
          {
            "update": 75,
            "environment_steps": 307200,
            "curriculum": 2,
            "completed_episodes": 1925,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.24143092334270477,
            "loss": 5.472808003425598,
            "approx_kl": 0.003326016902065021,
            "wall_seconds": 616.6524563999847
          },
          {
            "update": 76,
            "environment_steps": 311296,
            "curriculum": 3,
            "completed_episodes": 1930,
            "new_episodes": 5,
            "new_successes": 5,
            "mean_step_reward": -0.24655617773532867,
            "loss": 7.497823804616928,
            "approx_kl": 0.0023312318862735992,
            "wall_seconds": 625.8054316999915
          },
          {
            "update": 77,
            "environment_steps": 315392,
            "curriculum": 3,
            "completed_episodes": 1932,
            "new_episodes": 2,
            "new_successes": 2,
            "mean_step_reward": -0.2868633270263672,
            "loss": 2.927832640707493,
            "approx_kl": 0.0011651090153463883,
            "wall_seconds": 634.9607625999779
          },
          {
            "update": 78,
            "environment_steps": 319488,
            "curriculum": 3,
            "completed_episodes": 1932,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.3202403783798218,
            "loss": 0.9330675546079874,
            "approx_kl": 0.0038429762298619607,
            "wall_seconds": 644.2099426999921
          },
          {
            "update": 79,
            "environment_steps": 323584,
            "curriculum": 3,
            "completed_episodes": 1935,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.33342260122299194,
            "loss": 2.641499176621437,
            "approx_kl": 0.0034636732070794096,
            "wall_seconds": 653.8751434999867
          },
          {
            "update": 80,
            "environment_steps": 327680,
            "curriculum": 3,
            "completed_episodes": 1941,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.32841765880584717,
            "loss": 6.594112426042557,
            "approx_kl": 0.004077221319676028,
            "wall_seconds": 663.1527100999956
          },
          {
            "update": 81,
            "environment_steps": 331776,
            "curriculum": 3,
            "completed_episodes": 1943,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.31450921297073364,
            "loss": 2.6554088443517685,
            "approx_kl": 0.005129300512635382,
            "wall_seconds": 672.4615891999856
          },
          {
            "update": 82,
            "environment_steps": 335872,
            "curriculum": 3,
            "completed_episodes": 1951,
            "new_episodes": 8,
            "new_successes": 2,
            "mean_step_reward": -0.3160817623138428,
            "loss": 7.644163578748703,
            "approx_kl": 0.0040176297789003,
            "wall_seconds": 680.0834685000009
          },
          {
            "update": 83,
            "environment_steps": 339968,
            "curriculum": 3,
            "completed_episodes": 1957,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2664596736431122,
            "loss": 5.7937513291835785,
            "approx_kl": 0.0022228492071008077,
            "wall_seconds": 686.1952535999881
          },
          {
            "update": 84,
            "environment_steps": 344064,
            "curriculum": 3,
            "completed_episodes": 1963,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.29038870334625244,
            "loss": 5.664179593324661,
            "approx_kl": 0.003601250678912038,
            "wall_seconds": 694.0826415999909
          },
          {
            "update": 85,
            "environment_steps": 348160,
            "curriculum": 3,
            "completed_episodes": 1970,
            "new_episodes": 7,
            "new_successes": 3,
            "mean_step_reward": -0.26229798793792725,
            "loss": 6.641157954931259,
            "approx_kl": 0.001986314859095728,
            "wall_seconds": 702.0040498999879
          },
          {
            "update": 86,
            "environment_steps": 352256,
            "curriculum": 3,
            "completed_episodes": 1974,
            "new_episodes": 4,
            "new_successes": 3,
            "mean_step_reward": -0.25731754302978516,
            "loss": 4.601674169301987,
            "approx_kl": 0.004224866230288171,
            "wall_seconds": 709.8876768999908
          },
          {
            "update": 87,
            "environment_steps": 356352,
            "curriculum": 3,
            "completed_episodes": 1975,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.31975024938583374,
            "loss": 1.8260252103209496,
            "approx_kl": 0.002993184183651465,
            "wall_seconds": 717.6849738000019
          },
          {
            "update": 88,
            "environment_steps": 360448,
            "curriculum": 3,
            "completed_episodes": 1978,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3156084418296814,
            "loss": 4.156688705086708,
            "approx_kl": 0.0025180202665069373,
            "wall_seconds": 725.4680297999876
          },
          {
            "update": 89,
            "environment_steps": 364544,
            "curriculum": 3,
            "completed_episodes": 1982,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.29552915692329407,
            "loss": 5.154711544513702,
            "approx_kl": 0.003640157956397161,
            "wall_seconds": 733.2660491999995
          },
          {
            "update": 90,
            "environment_steps": 368640,
            "curriculum": 3,
            "completed_episodes": 1985,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.24675758183002472,
            "loss": 3.9505660086870193,
            "approx_kl": 0.002945547565104789,
            "wall_seconds": 741.0733375999844
          },
          {
            "update": 91,
            "environment_steps": 372736,
            "curriculum": 3,
            "completed_episodes": 1988,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2880622446537018,
            "loss": 4.736708447337151,
            "approx_kl": 0.002531453090341529,
            "wall_seconds": 748.9218567999778
          },
          {
            "update": 92,
            "environment_steps": 376832,
            "curriculum": 3,
            "completed_episodes": 1997,
            "new_episodes": 9,
            "new_successes": 0,
            "mean_step_reward": -0.320257306098938,
            "loss": 9.013528138399124,
            "approx_kl": 0.005056442123532179,
            "wall_seconds": 756.7807425999781
          },
          {
            "update": 93,
            "environment_steps": 380928,
            "curriculum": 3,
            "completed_episodes": 2000,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.28529131412506104,
            "loss": 4.184868887066841,
            "approx_kl": 0.003031849792023422,
            "wall_seconds": 764.695682599995
          },
          {
            "update": 94,
            "environment_steps": 385024,
            "curriculum": 3,
            "completed_episodes": 2002,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3273259997367859,
            "loss": 2.1535354554653168,
            "approx_kl": 0.0072774304207996465,
            "wall_seconds": 772.5669588999881
          },
          {
            "update": 95,
            "environment_steps": 389120,
            "curriculum": 3,
            "completed_episodes": 2008,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.28710341453552246,
            "loss": 6.9955220222473145,
            "approx_kl": 0.002749026602032245,
            "wall_seconds": 780.4895519999845
          },
          {
            "update": 96,
            "environment_steps": 393216,
            "curriculum": 3,
            "completed_episodes": 2010,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2762259244918823,
            "loss": 5.7297422885894775,
            "approx_kl": 0.002277579162182519,
            "wall_seconds": 788.4017186999845
          },
          {
            "update": 97,
            "environment_steps": 397312,
            "curriculum": 3,
            "completed_episodes": 2010,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.3229590952396393,
            "loss": 0.513784758746624,
            "approx_kl": 0.005801345918371226,
            "wall_seconds": 796.3520973999985
          },
          {
            "update": 98,
            "environment_steps": 401408,
            "curriculum": 3,
            "completed_episodes": 2013,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.30310776829719543,
            "loss": 4.191921576857567,
            "approx_kl": 0.010048853027910809,
            "wall_seconds": 804.325597099989
          },
          {
            "update": 99,
            "environment_steps": 405504,
            "curriculum": 3,
            "completed_episodes": 2018,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.29381924867630005,
            "loss": 6.675540253520012,
            "approx_kl": 0.0039825490184739465,
            "wall_seconds": 812.3507722999784
          },
          {
            "update": 100,
            "environment_steps": 409600,
            "curriculum": 3,
            "completed_episodes": 2021,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29791373014450073,
            "loss": 3.1890399903059006,
            "approx_kl": 0.0067919997145509114,
            "wall_seconds": 820.350920900004
          },
          {
            "update": 101,
            "environment_steps": 413696,
            "curriculum": 3,
            "completed_episodes": 2027,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.3085200786590576,
            "loss": 7.6454814076423645,
            "approx_kl": 0.0067806224560627015,
            "wall_seconds": 828.3638933000038
          },
          {
            "update": 102,
            "environment_steps": 417792,
            "curriculum": 3,
            "completed_episodes": 2033,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.28042325377464294,
            "loss": 7.574552625417709,
            "approx_kl": 0.002528538283513626,
            "wall_seconds": 836.3572633999866
          },
          {
            "update": 103,
            "environment_steps": 421888,
            "curriculum": 3,
            "completed_episodes": 2035,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.27395153045654297,
            "loss": 3.2598155587911606,
            "approx_kl": 0.003871512562909629,
            "wall_seconds": 844.3647401999915
          },
          {
            "update": 104,
            "environment_steps": 425984,
            "curriculum": 3,
            "completed_episodes": 2039,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.30991747975349426,
            "loss": 4.239519611001015,
            "approx_kl": 0.0038319032773870276,
            "wall_seconds": 852.3910423999769
          },
          {
            "update": 105,
            "environment_steps": 430080,
            "curriculum": 3,
            "completed_episodes": 2045,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2898380756378174,
            "loss": 5.910415500402451,
            "approx_kl": 0.0047065912276593735,
            "wall_seconds": 860.4072308999894
          },
          {
            "update": 106,
            "environment_steps": 434176,
            "curriculum": 3,
            "completed_episodes": 2046,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.32288211584091187,
            "loss": 1.8253993541002274,
            "approx_kl": 0.021343963147955947,
            "wall_seconds": 868.280340199999
          },
          {
            "update": 107,
            "environment_steps": 438272,
            "curriculum": 3,
            "completed_episodes": 2048,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30823057889938354,
            "loss": 3.841364771127701,
            "approx_kl": 0.0037957252588967094,
            "wall_seconds": 876.2591132999805
          },
          {
            "update": 108,
            "environment_steps": 442368,
            "curriculum": 3,
            "completed_episodes": 2052,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.27017155289649963,
            "loss": 5.6467482298612595,
            "approx_kl": 0.004166093403910054,
            "wall_seconds": 884.2638661999954
          },
          {
            "update": 109,
            "environment_steps": 446464,
            "curriculum": 3,
            "completed_episodes": 2056,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.25831443071365356,
            "loss": 6.7492329478263855,
            "approx_kl": 0.004386254786368227,
            "wall_seconds": 892.2630267999775
          },
          {
            "update": 110,
            "environment_steps": 450560,
            "curriculum": 3,
            "completed_episodes": 2062,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.2877730131149292,
            "loss": 6.492457389831543,
            "approx_kl": 0.004175447280431399,
            "wall_seconds": 900.2686960999854
          },
          {
            "update": 111,
            "environment_steps": 454656,
            "curriculum": 3,
            "completed_episodes": 2069,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2932347059249878,
            "loss": 6.837751477956772,
            "approx_kl": 0.0030137704397930065,
            "wall_seconds": 908.2343961999868
          },
          {
            "update": 112,
            "environment_steps": 458752,
            "curriculum": 3,
            "completed_episodes": 2071,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2754054069519043,
            "loss": 3.166309580206871,
            "approx_kl": 0.005712748048608773,
            "wall_seconds": 916.2585258999898
          },
          {
            "update": 113,
            "environment_steps": 462848,
            "curriculum": 3,
            "completed_episodes": 2072,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.29593417048454285,
            "loss": 2.2794054225087166,
            "approx_kl": 0.003587244615118834,
            "wall_seconds": 924.2566080999968
          },
          {
            "update": 114,
            "environment_steps": 466944,
            "curriculum": 3,
            "completed_episodes": 2078,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.3061821162700653,
            "loss": 8.661566823720932,
            "approx_kl": 0.00253665567470307,
            "wall_seconds": 932.19065579999
          },
          {
            "update": 115,
            "environment_steps": 471040,
            "curriculum": 3,
            "completed_episodes": 2080,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30590251088142395,
            "loss": 2.916888043284416,
            "approx_kl": 0.0024030456042964943,
            "wall_seconds": 940.1653850000002
          },
          {
            "update": 116,
            "environment_steps": 475136,
            "curriculum": 3,
            "completed_episodes": 2080,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.3234184980392456,
            "loss": 0.7061833292245865,
            "approx_kl": 0.0034057242755807238,
            "wall_seconds": 948.1831651999964
          },
          {
            "update": 117,
            "environment_steps": 479232,
            "curriculum": 3,
            "completed_episodes": 2083,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3217267394065857,
            "loss": 3.222271718084812,
            "approx_kl": 0.009902096397127025,
            "wall_seconds": 956.1822599000006
          },
          {
            "update": 118,
            "environment_steps": 483328,
            "curriculum": 3,
            "completed_episodes": 2089,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.311889111995697,
            "loss": 6.126597732305527,
            "approx_kl": 0.0042949217749992386,
            "wall_seconds": 962.0933506999863
          },
          {
            "update": 119,
            "environment_steps": 487424,
            "curriculum": 3,
            "completed_episodes": 2092,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29469162225723267,
            "loss": 4.554159879684448,
            "approx_kl": 0.0065378817434975645,
            "wall_seconds": 968.5254340999818
          },
          {
            "update": 120,
            "environment_steps": 491520,
            "curriculum": 3,
            "completed_episodes": 2101,
            "new_episodes": 9,
            "new_successes": 1,
            "mean_step_reward": -0.2931481599807739,
            "loss": 10.044893383979797,
            "approx_kl": 0.004303711659304099,
            "wall_seconds": 977.6365086000005
          },
          {
            "update": 121,
            "environment_steps": 495616,
            "curriculum": 3,
            "completed_episodes": 2108,
            "new_episodes": 7,
            "new_successes": 3,
            "mean_step_reward": -0.26587405800819397,
            "loss": 9.036986738443375,
            "approx_kl": 0.0035506938220351003,
            "wall_seconds": 989.787912699976
          },
          {
            "update": 122,
            "environment_steps": 499712,
            "curriculum": 3,
            "completed_episodes": 2109,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.27283933758735657,
            "loss": 2.2431324794888496,
            "approx_kl": 0.004905846235487843,
            "wall_seconds": 1001.5020845999825
          },
          {
            "update": 123,
            "environment_steps": 503808,
            "curriculum": 3,
            "completed_episodes": 2112,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2939125895500183,
            "loss": 5.4812783151865005,
            "approx_kl": 0.004718928157672053,
            "wall_seconds": 1014.2475139999879
          },
          {
            "update": 124,
            "environment_steps": 507904,
            "curriculum": 3,
            "completed_episodes": 2116,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2740291953086853,
            "loss": 4.263820201158524,
            "approx_kl": 0.005117899247125024,
            "wall_seconds": 1027.0176396000024
          },
          {
            "update": 125,
            "environment_steps": 512000,
            "curriculum": 3,
            "completed_episodes": 2117,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3174803555011749,
            "loss": 1.8328094184398651,
            "approx_kl": 0.005564369461353635,
            "wall_seconds": 1039.9323709000018
          },
          {
            "update": 126,
            "environment_steps": 516096,
            "curriculum": 3,
            "completed_episodes": 2120,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2990005612373352,
            "loss": 5.081752672791481,
            "approx_kl": 0.004660124004658428,
            "wall_seconds": 1052.852049499983
          },
          {
            "update": 127,
            "environment_steps": 520192,
            "curriculum": 3,
            "completed_episodes": 2127,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.2796427011489868,
            "loss": 9.858467906713486,
            "approx_kl": 0.004414684139192104,
            "wall_seconds": 1065.8338188999915
          },
          {
            "update": 128,
            "environment_steps": 524288,
            "curriculum": 3,
            "completed_episodes": 2131,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.26404786109924316,
            "loss": 7.334112703800201,
            "approx_kl": 0.00338583850498253,
            "wall_seconds": 1076.7076301999914
          },
          {
            "update": 129,
            "environment_steps": 528384,
            "curriculum": 3,
            "completed_episodes": 2138,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.2979546785354614,
            "loss": 7.143020778894424,
            "approx_kl": 0.0038489146772917593,
            "wall_seconds": 1084.902720699989
          },
          {
            "update": 130,
            "environment_steps": 532480,
            "curriculum": 3,
            "completed_episodes": 2141,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.30838504433631897,
            "loss": 3.8910164684057236,
            "approx_kl": 0.0024554208011977607,
            "wall_seconds": 1093.008998299978
          },
          {
            "update": 131,
            "environment_steps": 536576,
            "curriculum": 3,
            "completed_episodes": 2145,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.29483887553215027,
            "loss": 3.452658951282501,
            "approx_kl": 0.0025694476844364544,
            "wall_seconds": 1102.15300999998
          },
          {
            "update": 132,
            "environment_steps": 540672,
            "curriculum": 3,
            "completed_episodes": 2147,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3068563938140869,
            "loss": 2.6663206592202187,
            "approx_kl": 0.007261400534844142,
            "wall_seconds": 1111.3668622999976
          },
          {
            "update": 133,
            "environment_steps": 544768,
            "curriculum": 3,
            "completed_episodes": 2152,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2878630459308624,
            "loss": 5.962271004915237,
            "approx_kl": 0.008344307794686756,
            "wall_seconds": 1120.5642176999827
          },
          {
            "update": 134,
            "environment_steps": 548864,
            "curriculum": 3,
            "completed_episodes": 2153,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.27869540452957153,
            "loss": 2.6759223639965057,
            "approx_kl": 0.004842050351726357,
            "wall_seconds": 1129.8264994999918
          },
          {
            "update": 135,
            "environment_steps": 552960,
            "curriculum": 3,
            "completed_episodes": 2157,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.30950993299484253,
            "loss": 3.5047865360975266,
            "approx_kl": 0.003208632306268555,
            "wall_seconds": 1139.1230127000017
          },
          {
            "update": 136,
            "environment_steps": 557056,
            "curriculum": 3,
            "completed_episodes": 2162,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.28283873200416565,
            "loss": 7.078308701515198,
            "approx_kl": 0.0037872690700169187,
            "wall_seconds": 1148.4659630999959
          },
          {
            "update": 137,
            "environment_steps": 561152,
            "curriculum": 3,
            "completed_episodes": 2168,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.2599653899669647,
            "loss": 8.951664000749588,
            "approx_kl": 0.003603160657803528,
            "wall_seconds": 1157.190516999981
          },
          {
            "update": 138,
            "environment_steps": 565248,
            "curriculum": 3,
            "completed_episodes": 2173,
            "new_episodes": 5,
            "new_successes": 3,
            "mean_step_reward": -0.23806720972061157,
            "loss": 9.522816628217697,
            "approx_kl": 0.0021573978428932605,
            "wall_seconds": 1163.3128726999857
          },
          {
            "update": 139,
            "environment_steps": 569344,
            "curriculum": 3,
            "completed_episodes": 2180,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.2663973569869995,
            "loss": 8.106617718935013,
            "approx_kl": 0.0038795288255641935,
            "wall_seconds": 1171.2475306999986
          },
          {
            "update": 140,
            "environment_steps": 573440,
            "curriculum": 3,
            "completed_episodes": 2184,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.28799623250961304,
            "loss": 4.498796299099922,
            "approx_kl": 0.0031675900190748507,
            "wall_seconds": 1179.1587056999851
          },
          {
            "update": 141,
            "environment_steps": 577536,
            "curriculum": 3,
            "completed_episodes": 2187,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2960127890110016,
            "loss": 5.384368926286697,
            "approx_kl": 0.0026773695226438576,
            "wall_seconds": 1186.9750953000039
          },
          {
            "update": 142,
            "environment_steps": 581632,
            "curriculum": 3,
            "completed_episodes": 2192,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.30294549465179443,
            "loss": 5.87361016869545,
            "approx_kl": 0.002724448577282601,
            "wall_seconds": 1195.0231266999908
          },
          {
            "update": 143,
            "environment_steps": 585728,
            "curriculum": 3,
            "completed_episodes": 2195,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2557690143585205,
            "loss": 5.062629908323288,
            "approx_kl": 0.002916908899351256,
            "wall_seconds": 1202.9764224999817
          },
          {
            "update": 144,
            "environment_steps": 589824,
            "curriculum": 3,
            "completed_episodes": 2196,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.2954370975494385,
            "loss": 1.8868382722139359,
            "approx_kl": 0.0037053394225949887,
            "wall_seconds": 1210.8012589999998
          },
          {
            "update": 145,
            "environment_steps": 593920,
            "curriculum": 3,
            "completed_episodes": 2198,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30470597743988037,
            "loss": 3.0980828255414963,
            "approx_kl": 0.004651207675124169,
            "wall_seconds": 1218.616200999997
          },
          {
            "update": 146,
            "environment_steps": 598016,
            "curriculum": 3,
            "completed_episodes": 2204,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.3100120723247528,
            "loss": 7.398072421550751,
            "approx_kl": 0.004556144089292502,
            "wall_seconds": 1226.4468796
          },
          {
            "update": 147,
            "environment_steps": 602112,
            "curriculum": 3,
            "completed_episodes": 2208,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.310375452041626,
            "loss": 4.696838393807411,
            "approx_kl": 0.005494559198268689,
            "wall_seconds": 1234.32605839998
          },
          {
            "update": 148,
            "environment_steps": 606208,
            "curriculum": 3,
            "completed_episodes": 2215,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.29372912645339966,
            "loss": 7.866695314645767,
            "approx_kl": 0.0027463485675980337,
            "wall_seconds": 1242.3219341999793
          },
          {
            "update": 149,
            "environment_steps": 610304,
            "curriculum": 3,
            "completed_episodes": 2218,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2508687376976013,
            "loss": 4.270640671253204,
            "approx_kl": 0.004612743370671524,
            "wall_seconds": 1250.4231222999806
          },
          {
            "update": 150,
            "environment_steps": 614400,
            "curriculum": 3,
            "completed_episodes": 2221,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.25829464197158813,
            "loss": 5.706988960504532,
            "approx_kl": 0.004869681686614058,
            "wall_seconds": 1258.4762031999999
          },
          {
            "update": 151,
            "environment_steps": 618496,
            "curriculum": 3,
            "completed_episodes": 2224,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2849491238594055,
            "loss": 4.430895879864693,
            "approx_kl": 0.00387602271985088,
            "wall_seconds": 1266.6072417999967
          },
          {
            "update": 152,
            "environment_steps": 622592,
            "curriculum": 3,
            "completed_episodes": 2231,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.23223942518234253,
            "loss": 8.929513335227966,
            "approx_kl": 0.0042525552125880495,
            "wall_seconds": 1274.7558008999913
          },
          {
            "update": 153,
            "environment_steps": 626688,
            "curriculum": 3,
            "completed_episodes": 2234,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.24901258945465088,
            "loss": 4.3373768627643585,
            "approx_kl": 0.0032863841970538488,
            "wall_seconds": 1282.9100456000015
          },
          {
            "update": 154,
            "environment_steps": 630784,
            "curriculum": 3,
            "completed_episodes": 2237,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2873271405696869,
            "loss": 4.578722700476646,
            "approx_kl": 0.0038442707518697716,
            "wall_seconds": 1291.0477712999855
          },
          {
            "update": 155,
            "environment_steps": 634880,
            "curriculum": 3,
            "completed_episodes": 2240,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29206526279449463,
            "loss": 3.3861503154039383,
            "approx_kl": 0.004869506834438653,
            "wall_seconds": 1299.2024651999818
          },
          {
            "update": 156,
            "environment_steps": 638976,
            "curriculum": 3,
            "completed_episodes": 2247,
            "new_episodes": 7,
            "new_successes": 3,
            "mean_step_reward": -0.24821233749389648,
            "loss": 10.066505581140518,
            "approx_kl": 0.0016401618668169249,
            "wall_seconds": 1307.3221432999999
          },
          {
            "update": 157,
            "environment_steps": 643072,
            "curriculum": 3,
            "completed_episodes": 2253,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.29561394453048706,
            "loss": 6.6701400578022,
            "approx_kl": 0.005311655182595132,
            "wall_seconds": 1315.4166217999882
          },
          {
            "update": 158,
            "environment_steps": 647168,
            "curriculum": 3,
            "completed_episodes": 2255,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2705272436141968,
            "loss": 1.9737870544195175,
            "approx_kl": 0.003912028796548839,
            "wall_seconds": 1323.4454739999783
          },
          {
            "update": 159,
            "environment_steps": 651264,
            "curriculum": 3,
            "completed_episodes": 2259,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.28771311044692993,
            "loss": 5.381802022457123,
            "approx_kl": 0.005530900658413884,
            "wall_seconds": 1331.5522713999962
          },
          {
            "update": 160,
            "environment_steps": 655360,
            "curriculum": 3,
            "completed_episodes": 2262,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2800248861312866,
            "loss": 3.2658418715000153,
            "approx_kl": 0.005855196604898083,
            "wall_seconds": 1339.590107600001
          },
          {
            "update": 161,
            "environment_steps": 659456,
            "curriculum": 3,
            "completed_episodes": 2270,
            "new_episodes": 8,
            "new_successes": 0,
            "mean_step_reward": -0.2906333804130554,
            "loss": 9.026658743619919,
            "approx_kl": 0.0037245548537612194,
            "wall_seconds": 1347.722786999977
          },
          {
            "update": 162,
            "environment_steps": 663552,
            "curriculum": 3,
            "completed_episodes": 2272,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2371402084827423,
            "loss": 4.431335434317589,
            "approx_kl": 0.004843538121349411,
            "wall_seconds": 1355.8560598999902
          },
          {
            "update": 163,
            "environment_steps": 667648,
            "curriculum": 3,
            "completed_episodes": 2275,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.27346453070640564,
            "loss": 5.699747771024704,
            "approx_kl": 0.003587903987863683,
            "wall_seconds": 1363.970696399978
          },
          {
            "update": 164,
            "environment_steps": 671744,
            "curriculum": 3,
            "completed_episodes": 2276,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.229533851146698,
            "loss": 5.264358222484589,
            "approx_kl": 0.003959328469136381,
            "wall_seconds": 1372.1874964000017
          },
          {
            "update": 165,
            "environment_steps": 675840,
            "curriculum": 3,
            "completed_episodes": 2279,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2758665084838867,
            "loss": 4.149466276168823,
            "approx_kl": 0.002898589318647282,
            "wall_seconds": 1380.3170988999773
          },
          {
            "update": 166,
            "environment_steps": 679936,
            "curriculum": 3,
            "completed_episodes": 2282,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2609789967536926,
            "loss": 3.960729330778122,
            "approx_kl": 0.002579346206402988,
            "wall_seconds": 1388.4212804999843
          },
          {
            "update": 167,
            "environment_steps": 684032,
            "curriculum": 3,
            "completed_episodes": 2288,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.2740776538848877,
            "loss": 7.148020923137665,
            "approx_kl": 0.004354099079137086,
            "wall_seconds": 1396.6298752999865
          },
          {
            "update": 168,
            "environment_steps": 688128,
            "curriculum": 3,
            "completed_episodes": 2289,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.2690277099609375,
            "loss": 2.5072687193751335,
            "approx_kl": 0.0040860726694518235,
            "wall_seconds": 1404.7862958000042
          },
          {
            "update": 169,
            "environment_steps": 692224,
            "curriculum": 3,
            "completed_episodes": 2295,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.28300780057907104,
            "loss": 6.4414728581905365,
            "approx_kl": 0.004802311181265395,
            "wall_seconds": 1412.8310029999993
          },
          {
            "update": 170,
            "environment_steps": 696320,
            "curriculum": 3,
            "completed_episodes": 2301,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.2827276587486267,
            "loss": 7.302999675273895,
            "approx_kl": 0.005283993034026935,
            "wall_seconds": 1420.9903407999955
          },
          {
            "update": 171,
            "environment_steps": 700416,
            "curriculum": 3,
            "completed_episodes": 2305,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3046152591705322,
            "loss": 4.930046707391739,
            "approx_kl": 0.0034867418798967265,
            "wall_seconds": 1428.499015199981
          },
          {
            "update": 172,
            "environment_steps": 704512,
            "curriculum": 3,
            "completed_episodes": 2307,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2725744843482971,
            "loss": 5.126876100897789,
            "approx_kl": 0.004155265705776401,
            "wall_seconds": 1434.3349150999857
          },
          {
            "update": 173,
            "environment_steps": 708608,
            "curriculum": 3,
            "completed_episodes": 2310,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29352396726608276,
            "loss": 3.3663389682769775,
            "approx_kl": 0.0059363326745369704,
            "wall_seconds": 1441.5732526999782
          },
          {
            "update": 174,
            "environment_steps": 712704,
            "curriculum": 3,
            "completed_episodes": 2313,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2813059091567993,
            "loss": 5.854391798377037,
            "approx_kl": 0.003495969998766668,
            "wall_seconds": 1450.798364599992
          },
          {
            "update": 175,
            "environment_steps": 716800,
            "curriculum": 3,
            "completed_episodes": 2319,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2536834478378296,
            "loss": 7.539875507354736,
            "approx_kl": 0.003796489154410665,
            "wall_seconds": 1458.2749899000046
          },
          {
            "update": 176,
            "environment_steps": 720896,
            "curriculum": 3,
            "completed_episodes": 2326,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2874464988708496,
            "loss": 8.252563834190369,
            "approx_kl": 0.0027078637958766194,
            "wall_seconds": 1467.0306071999948
          },
          {
            "update": 177,
            "environment_steps": 724992,
            "curriculum": 3,
            "completed_episodes": 2327,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.30013906955718994,
            "loss": 2.6273383423686028,
            "approx_kl": 0.004900776506474358,
            "wall_seconds": 1476.1960382999969
          },
          {
            "update": 178,
            "environment_steps": 729088,
            "curriculum": 3,
            "completed_episodes": 2329,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2805441915988922,
            "loss": 3.444340795278549,
            "approx_kl": 0.010098863447637996,
            "wall_seconds": 1485.4081123999786
          },
          {
            "update": 179,
            "environment_steps": 733184,
            "curriculum": 3,
            "completed_episodes": 2333,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.285548597574234,
            "loss": 4.11740942299366,
            "approx_kl": 0.0096657893864176,
            "wall_seconds": 1494.6537853000045
          },
          {
            "update": 180,
            "environment_steps": 737280,
            "curriculum": 3,
            "completed_episodes": 2341,
            "new_episodes": 8,
            "new_successes": 2,
            "mean_step_reward": -0.22842420637607574,
            "loss": 8.584440141916275,
            "approx_kl": 0.004580406239256263,
            "wall_seconds": 1503.9834969999793
          },
          {
            "update": 181,
            "environment_steps": 741376,
            "curriculum": 3,
            "completed_episodes": 2347,
            "new_episodes": 6,
            "new_successes": 3,
            "mean_step_reward": -0.20351412892341614,
            "loss": 7.491285502910614,
            "approx_kl": 0.0018119264150300296,
            "wall_seconds": 1513.4226818000025
          },
          {
            "update": 182,
            "environment_steps": 745472,
            "curriculum": 3,
            "completed_episodes": 2350,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2683979868888855,
            "loss": 5.095632463693619,
            "approx_kl": 0.0022080607905081706,
            "wall_seconds": 1522.828116299992
          },
          {
            "update": 183,
            "environment_steps": 749568,
            "curriculum": 3,
            "completed_episodes": 2351,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3279823064804077,
            "loss": 1.5672529488801956,
            "approx_kl": 0.004713193648058223,
            "wall_seconds": 1530.4212926999899
          },
          {
            "update": 184,
            "environment_steps": 753664,
            "curriculum": 3,
            "completed_episodes": 2357,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2745078206062317,
            "loss": 6.056465283036232,
            "approx_kl": 0.004844576364121167,
            "wall_seconds": 1538.864146899985
          },
          {
            "update": 185,
            "environment_steps": 757760,
            "curriculum": 3,
            "completed_episodes": 2362,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2891201972961426,
            "loss": 4.382366985082626,
            "approx_kl": 0.0016733403463149443,
            "wall_seconds": 1547.9561685999797
          },
          {
            "update": 186,
            "environment_steps": 761856,
            "curriculum": 3,
            "completed_episodes": 2366,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2993112802505493,
            "loss": 3.0169133245944977,
            "approx_kl": 0.0032409213526989333,
            "wall_seconds": 1557.1536305999907
          },
          {
            "update": 187,
            "environment_steps": 765952,
            "curriculum": 3,
            "completed_episodes": 2367,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.3017955422401428,
            "loss": 1.7257517725229263,
            "approx_kl": 0.00321704782982124,
            "wall_seconds": 1566.3348511999939
          },
          {
            "update": 188,
            "environment_steps": 770048,
            "curriculum": 3,
            "completed_episodes": 2370,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.32581549882888794,
            "loss": 3.2952835261821747,
            "approx_kl": 0.002353099238462164,
            "wall_seconds": 1575.581173499988
          },
          {
            "update": 189,
            "environment_steps": 774144,
            "curriculum": 3,
            "completed_episodes": 2378,
            "new_episodes": 8,
            "new_successes": 1,
            "mean_step_reward": -0.32140052318573,
            "loss": 8.840545535087585,
            "approx_kl": 0.005220874463702785,
            "wall_seconds": 1584.9121358999982
          },
          {
            "update": 190,
            "environment_steps": 778240,
            "curriculum": 3,
            "completed_episodes": 2383,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.25820547342300415,
            "loss": 5.940881371498108,
            "approx_kl": 0.003920930650565424,
            "wall_seconds": 1594.2017228999757
          },
          {
            "update": 191,
            "environment_steps": 782336,
            "curriculum": 3,
            "completed_episodes": 2387,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.25746777653694153,
            "loss": 5.652920350432396,
            "approx_kl": 0.002505917385860812,
            "wall_seconds": 1602.3808800999832
          },
          {
            "update": 192,
            "environment_steps": 786432,
            "curriculum": 3,
            "completed_episodes": 2391,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2804870009422302,
            "loss": 5.7707546800374985,
            "approx_kl": 0.0038502476818393916,
            "wall_seconds": 1608.685964299977
          },
          {
            "update": 193,
            "environment_steps": 790528,
            "curriculum": 3,
            "completed_episodes": 2392,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.3058580458164215,
            "loss": 3.711888909339905,
            "approx_kl": 0.006740500695741503,
            "wall_seconds": 1616.4931367999816
          },
          {
            "update": 194,
            "environment_steps": 794624,
            "curriculum": 3,
            "completed_episodes": 2396,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.31379422545433044,
            "loss": 5.049549028277397,
            "approx_kl": 0.004044523773700348,
            "wall_seconds": 1624.4614715999924
          },
          {
            "update": 195,
            "environment_steps": 798720,
            "curriculum": 3,
            "completed_episodes": 2403,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.2973777651786804,
            "loss": 6.947379678487778,
            "approx_kl": 0.003920755290891975,
            "wall_seconds": 1632.334439999977
          },
          {
            "update": 196,
            "environment_steps": 802816,
            "curriculum": 3,
            "completed_episodes": 2406,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.261684775352478,
            "loss": 6.35332116484642,
            "approx_kl": 0.004210955028611352,
            "wall_seconds": 1640.189522200002
          },
          {
            "update": 197,
            "environment_steps": 806912,
            "curriculum": 3,
            "completed_episodes": 2409,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.26342734694480896,
            "loss": 3.401605471968651,
            "approx_kl": 0.004345555978943594,
            "wall_seconds": 1647.9646061999956
          },
          {
            "update": 198,
            "environment_steps": 811008,
            "curriculum": 3,
            "completed_episodes": 2413,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.30374276638031006,
            "loss": 5.016822457313538,
            "approx_kl": 0.004256709940818837,
            "wall_seconds": 1655.69048379999
          },
          {
            "update": 199,
            "environment_steps": 815104,
            "curriculum": 3,
            "completed_episodes": 2418,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.3010447323322296,
            "loss": 5.6324638575315475,
            "approx_kl": 0.005440640341475955,
            "wall_seconds": 1663.4876768999966
          },
          {
            "update": 200,
            "environment_steps": 819200,
            "curriculum": 3,
            "completed_episodes": 2422,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2898339629173279,
            "loss": 3.2053838968276978,
            "approx_kl": 0.0019615634737419896,
            "wall_seconds": 1671.2094933000044
          },
          {
            "update": 201,
            "environment_steps": 823296,
            "curriculum": 3,
            "completed_episodes": 2428,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.2805703282356262,
            "loss": 9.218411833047867,
            "approx_kl": 0.004954700227244757,
            "wall_seconds": 1679.1434426999767
          },
          {
            "update": 202,
            "environment_steps": 827392,
            "curriculum": 3,
            "completed_episodes": 2429,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.2815483510494232,
            "loss": 2.013062410056591,
            "approx_kl": 0.0033241253022424644,
            "wall_seconds": 1687.011395199981
          },
          {
            "update": 203,
            "environment_steps": 831488,
            "curriculum": 3,
            "completed_episodes": 2432,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3198404908180237,
            "loss": 3.6210290044546127,
            "approx_kl": 0.005007513831515098,
            "wall_seconds": 1695.0159572999983
          },
          {
            "update": 204,
            "environment_steps": 835584,
            "curriculum": 3,
            "completed_episodes": 2438,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.3196825087070465,
            "loss": 6.3419168293476105,
            "approx_kl": 0.0038250122361205285,
            "wall_seconds": 1703.0914718999993
          },
          {
            "update": 205,
            "environment_steps": 839680,
            "curriculum": 3,
            "completed_episodes": 2440,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.28530997037887573,
            "loss": 2.8893075957894325,
            "approx_kl": 0.004895734808087582,
            "wall_seconds": 1711.1451992999937
          },
          {
            "update": 206,
            "environment_steps": 843776,
            "curriculum": 3,
            "completed_episodes": 2443,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3081696331501007,
            "loss": 3.6294554471969604,
            "approx_kl": 0.005603259596682619,
            "wall_seconds": 1719.2679151999764
          },
          {
            "update": 207,
            "environment_steps": 847872,
            "curriculum": 3,
            "completed_episodes": 2445,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30402928590774536,
            "loss": 2.9869320765137672,
            "approx_kl": 0.002763275040706503,
            "wall_seconds": 1727.2281799999764
          },
          {
            "update": 208,
            "environment_steps": 851968,
            "curriculum": 3,
            "completed_episodes": 2449,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2940051555633545,
            "loss": 4.802380934357643,
            "approx_kl": 0.005053315202530939,
            "wall_seconds": 1735.3714173999906
          },
          {
            "update": 209,
            "environment_steps": 856064,
            "curriculum": 3,
            "completed_episodes": 2454,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.32199013233184814,
            "loss": 5.460069701075554,
            "approx_kl": 0.0035139079027430853,
            "wall_seconds": 1743.543351100001
          },
          {
            "update": 210,
            "environment_steps": 860160,
            "curriculum": 3,
            "completed_episodes": 2458,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.32029426097869873,
            "loss": 4.034124106168747,
            "approx_kl": 0.0065412296335125575,
            "wall_seconds": 1751.728537499992
          },
          {
            "update": 211,
            "environment_steps": 864256,
            "curriculum": 3,
            "completed_episodes": 2463,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.2715749740600586,
            "loss": 10.23474046587944,
            "approx_kl": 0.010145374888452352,
            "wall_seconds": 1759.9781289999955
          },
          {
            "update": 212,
            "environment_steps": 868352,
            "curriculum": 3,
            "completed_episodes": 2463,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.3090859651565552,
            "loss": 0.6846150904893875,
            "approx_kl": 0.0025849841731542256,
            "wall_seconds": 1768.1122292999935
          },
          {
            "update": 213,
            "environment_steps": 872448,
            "curriculum": 3,
            "completed_episodes": 2467,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.29995572566986084,
            "loss": 4.549734473228455,
            "approx_kl": 0.003404668237635633,
            "wall_seconds": 1776.3062494999904
          },
          {
            "update": 214,
            "environment_steps": 876544,
            "curriculum": 3,
            "completed_episodes": 2474,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.3177517056465149,
            "loss": 7.460210382938385,
            "approx_kl": 0.003117186008239514,
            "wall_seconds": 1784.5095299999812
          },
          {
            "update": 215,
            "environment_steps": 880640,
            "curriculum": 3,
            "completed_episodes": 2476,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2813491225242615,
            "loss": 2.905003510415554,
            "approx_kl": 0.004303413570596604,
            "wall_seconds": 1792.69719549999
          },
          {
            "update": 216,
            "environment_steps": 884736,
            "curriculum": 3,
            "completed_episodes": 2477,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.31229984760284424,
            "loss": 2.165698453783989,
            "approx_kl": 0.003005871803907212,
            "wall_seconds": 1800.8894392999937
          },
          {
            "update": 217,
            "environment_steps": 888832,
            "curriculum": 3,
            "completed_episodes": 2481,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.31705009937286377,
            "loss": 4.053733244538307,
            "approx_kl": 0.003615849796915427,
            "wall_seconds": 1809.0250172999804
          },
          {
            "update": 218,
            "environment_steps": 892928,
            "curriculum": 3,
            "completed_episodes": 2489,
            "new_episodes": 8,
            "new_successes": 0,
            "mean_step_reward": -0.32140129804611206,
            "loss": 8.158277332782745,
            "approx_kl": 0.0015919266897981288,
            "wall_seconds": 1817.2249918000016
          },
          {
            "update": 219,
            "environment_steps": 897024,
            "curriculum": 3,
            "completed_episodes": 2491,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2861122488975525,
            "loss": 3.8064177185297012,
            "approx_kl": 0.006793641172407661,
            "wall_seconds": 1825.3246925999993
          },
          {
            "update": 220,
            "environment_steps": 901120,
            "curriculum": 3,
            "completed_episodes": 2495,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.28710848093032837,
            "loss": 4.879373177886009,
            "approx_kl": 0.00592261915517156,
            "wall_seconds": 1833.4688051999838
          },
          {
            "update": 221,
            "environment_steps": 905216,
            "curriculum": 3,
            "completed_episodes": 2497,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2985028326511383,
            "loss": 4.470311656594276,
            "approx_kl": 0.004397801812956459,
            "wall_seconds": 1841.5670190999808
          },
          {
            "update": 222,
            "environment_steps": 909312,
            "curriculum": 3,
            "completed_episodes": 2501,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.32628363370895386,
            "loss": 3.6813880875706673,
            "approx_kl": 0.00311188846717414,
            "wall_seconds": 1849.6617868999892
          },
          {
            "update": 223,
            "environment_steps": 913408,
            "curriculum": 3,
            "completed_episodes": 2507,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.3189261257648468,
            "loss": 5.241933286190033,
            "approx_kl": 0.004004265470939572,
            "wall_seconds": 1857.8198582999757
          },
          {
            "update": 224,
            "environment_steps": 917504,
            "curriculum": 3,
            "completed_episodes": 2510,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.28390777111053467,
            "loss": 3.6053301841020584,
            "approx_kl": 0.005977418179099914,
            "wall_seconds": 1865.9893649999867
          },
          {
            "update": 225,
            "environment_steps": 921600,
            "curriculum": 3,
            "completed_episodes": 2512,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2969156503677368,
            "loss": 2.6856247410178185,
            "approx_kl": 0.008253281323050032,
            "wall_seconds": 1872.6111570999783
          },
          {
            "update": 226,
            "environment_steps": 925696,
            "curriculum": 3,
            "completed_episodes": 2515,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.28966113924980164,
            "loss": 3.0640496015548706,
            "approx_kl": 0.004610776515619364,
            "wall_seconds": 1878.3796161999926
          },
          {
            "update": 227,
            "environment_steps": 929792,
            "curriculum": 3,
            "completed_episodes": 2522,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.2460627257823944,
            "loss": 11.045032143592834,
            "approx_kl": 0.0025177435545629123,
            "wall_seconds": 1886.7600094999943
          },
          {
            "update": 228,
            "environment_steps": 933888,
            "curriculum": 3,
            "completed_episodes": 2526,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.21187841892242432,
            "loss": 7.581584244966507,
            "approx_kl": 0.0022241560527618276,
            "wall_seconds": 1895.966060799983
          },
          {
            "update": 229,
            "environment_steps": 937984,
            "curriculum": 3,
            "completed_episodes": 2530,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.24847477674484253,
            "loss": 7.834407448768616,
            "approx_kl": 0.0011047990246879635,
            "wall_seconds": 1903.0010019000038
          },
          {
            "update": 230,
            "environment_steps": 942080,
            "curriculum": 3,
            "completed_episodes": 2535,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.3006519675254822,
            "loss": 6.3613835871219635,
            "approx_kl": 0.0036233133978385013,
            "wall_seconds": 1912.1389635999803
          },
          {
            "update": 231,
            "environment_steps": 946176,
            "curriculum": 3,
            "completed_episodes": 2538,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.31231045722961426,
            "loss": 3.635674148797989,
            "approx_kl": 0.002335796489205677,
            "wall_seconds": 1921.2766509999929
          },
          {
            "update": 232,
            "environment_steps": 950272,
            "curriculum": 3,
            "completed_episodes": 2543,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2894103229045868,
            "loss": 4.665168151259422,
            "approx_kl": 0.003750570422198507,
            "wall_seconds": 1930.5021582999907
          },
          {
            "update": 233,
            "environment_steps": 954368,
            "curriculum": 3,
            "completed_episodes": 2545,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2848508656024933,
            "loss": 3.550499990582466,
            "approx_kl": 0.0030173034210747574,
            "wall_seconds": 1939.8004624999885
          },
          {
            "update": 234,
            "environment_steps": 958464,
            "curriculum": 3,
            "completed_episodes": 2547,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.32854360342025757,
            "loss": 1.7227217443287373,
            "approx_kl": 0.004686350106567261,
            "wall_seconds": 1949.0607489999966
          },
          {
            "update": 235,
            "environment_steps": 962560,
            "curriculum": 3,
            "completed_episodes": 2549,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.31838640570640564,
            "loss": 2.519077032804489,
            "approx_kl": 0.007357974307524273,
            "wall_seconds": 1958.367357799987
          },
          {
            "update": 236,
            "environment_steps": 966656,
            "curriculum": 3,
            "completed_episodes": 2553,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.32264620065689087,
            "loss": 4.422892227768898,
            "approx_kl": 0.004825408843316836,
            "wall_seconds": 1967.6747902999923
          },
          {
            "update": 237,
            "environment_steps": 970752,
            "curriculum": 3,
            "completed_episodes": 2559,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.31605395674705505,
            "loss": 6.259493052959442,
            "approx_kl": 0.00520429759126273,
            "wall_seconds": 1974.8054288999992
          },
          {
            "update": 238,
            "environment_steps": 974848,
            "curriculum": 3,
            "completed_episodes": 2562,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.30867937207221985,
            "loss": 3.861695498228073,
            "approx_kl": 0.0038792501472926233,
            "wall_seconds": 1983.9113928999868
          },
          {
            "update": 239,
            "environment_steps": 978944,
            "curriculum": 3,
            "completed_episodes": 2565,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2888846695423126,
            "loss": 4.104306578636169,
            "approx_kl": 0.005673035728250397,
            "wall_seconds": 1993.0946313999884
          },
          {
            "update": 240,
            "environment_steps": 983040,
            "curriculum": 3,
            "completed_episodes": 2567,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2671544551849365,
            "loss": 2.6087519228458405,
            "approx_kl": 0.005753011202614289,
            "wall_seconds": 2002.346470599994
          },
          {
            "update": 241,
            "environment_steps": 987136,
            "curriculum": 3,
            "completed_episodes": 2571,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2774620056152344,
            "loss": 5.608838766813278,
            "approx_kl": 0.0065462883649161085,
            "wall_seconds": 2011.6570161999844
          },
          {
            "update": 242,
            "environment_steps": 991232,
            "curriculum": 3,
            "completed_episodes": 2575,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.27549847960472107,
            "loss": 5.313737481832504,
            "approx_kl": 0.0043358866751077585,
            "wall_seconds": 2021.018011499982
          },
          {
            "update": 243,
            "environment_steps": 995328,
            "curriculum": 3,
            "completed_episodes": 2581,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2852122485637665,
            "loss": 8.166981279850006,
            "approx_kl": 0.0024943736389104743,
            "wall_seconds": 2030.3205044999777
          },
          {
            "update": 244,
            "environment_steps": 999424,
            "curriculum": 3,
            "completed_episodes": 2582,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.2920870780944824,
            "loss": 1.9781609177589417,
            "approx_kl": 0.003840194080112269,
            "wall_seconds": 2039.6394717999792
          },
          {
            "update": 245,
            "environment_steps": 1003520,
            "curriculum": 3,
            "completed_episodes": 2586,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.32598602771759033,
            "loss": 4.974181771278381,
            "approx_kl": 0.004260156179952901,
            "wall_seconds": 2046.51936809998
          },
          {
            "update": 246,
            "environment_steps": 1007616,
            "curriculum": 3,
            "completed_episodes": 2595,
            "new_episodes": 9,
            "new_successes": 1,
            "mean_step_reward": -0.30926015973091125,
            "loss": 10.63908365368843,
            "approx_kl": 0.0019519986271916423,
            "wall_seconds": 2052.304698199994
          },
          {
            "update": 247,
            "environment_steps": 1011712,
            "curriculum": 3,
            "completed_episodes": 2599,
            "new_episodes": 4,
            "new_successes": 3,
            "mean_step_reward": -0.22572660446166992,
            "loss": 8.06578516960144,
            "approx_kl": 0.0033282780314038973,
            "wall_seconds": 2058.1328378999897
          },
          {
            "update": 248,
            "environment_steps": 1015808,
            "curriculum": 3,
            "completed_episodes": 2604,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2575705647468567,
            "loss": 8.55118590593338,
            "approx_kl": 0.0013901168640586548,
            "wall_seconds": 2063.8975130000035
          },
          {
            "update": 249,
            "environment_steps": 1019904,
            "curriculum": 3,
            "completed_episodes": 2608,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.24747991561889648,
            "loss": 6.093246668577194,
            "approx_kl": 0.0035511867681634612,
            "wall_seconds": 2069.7105742999993
          },
          {
            "update": 250,
            "environment_steps": 1024000,
            "curriculum": 3,
            "completed_episodes": 2610,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3187532126903534,
            "loss": 3.34814815223217,
            "approx_kl": 0.004290093493182212,
            "wall_seconds": 2075.3958117
          },
          {
            "update": 251,
            "environment_steps": 1028096,
            "curriculum": 3,
            "completed_episodes": 2613,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2988532483577728,
            "loss": 3.8996367752552032,
            "approx_kl": 0.00388525380185456,
            "wall_seconds": 2081.1409115999995
          },
          {
            "update": 252,
            "environment_steps": 1032192,
            "curriculum": 3,
            "completed_episodes": 2620,
            "new_episodes": 7,
            "new_successes": 3,
            "mean_step_reward": -0.28821122646331787,
            "loss": 7.364744454622269,
            "approx_kl": 0.001783748510206351,
            "wall_seconds": 2086.983964199986
          },
          {
            "update": 253,
            "environment_steps": 1036288,
            "curriculum": 3,
            "completed_episodes": 2625,
            "new_episodes": 5,
            "new_successes": 3,
            "mean_step_reward": -0.2634434700012207,
            "loss": 5.507377594709396,
            "approx_kl": 0.003924985157937044,
            "wall_seconds": 2092.736211099982
          },
          {
            "update": 254,
            "environment_steps": 1040384,
            "curriculum": 3,
            "completed_episodes": 2629,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.30872786045074463,
            "loss": 4.1119585037231445,
            "approx_kl": 0.0034498220193199813,
            "wall_seconds": 2098.474759899982
          },
          {
            "update": 255,
            "environment_steps": 1044480,
            "curriculum": 3,
            "completed_episodes": 2633,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3259216248989105,
            "loss": 4.316209360957146,
            "approx_kl": 0.0038563286761927884,
            "wall_seconds": 2104.2003043000004
          },
          {
            "update": 256,
            "environment_steps": 1048576,
            "curriculum": 3,
            "completed_episodes": 2637,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2954927682876587,
            "loss": 4.735881790518761,
            "approx_kl": 0.0037722937668149825,
            "wall_seconds": 2109.987274199986
          },
          {
            "update": 257,
            "environment_steps": 1052672,
            "curriculum": 3,
            "completed_episodes": 2640,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2700352072715759,
            "loss": 4.329792872071266,
            "approx_kl": 0.0031349894306913484,
            "wall_seconds": 2115.7721899999888
          },
          {
            "update": 258,
            "environment_steps": 1056768,
            "curriculum": 3,
            "completed_episodes": 2646,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.28755807876586914,
            "loss": 8.51945686340332,
            "approx_kl": 0.004038929448142881,
            "wall_seconds": 2121.5561283999996
          },
          {
            "update": 259,
            "environment_steps": 1060864,
            "curriculum": 3,
            "completed_episodes": 2647,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.29938775300979614,
            "loss": 2.357355624437332,
            "approx_kl": 0.004193669719825266,
            "wall_seconds": 2127.39868329998
          },
          {
            "update": 260,
            "environment_steps": 1064960,
            "curriculum": 3,
            "completed_episodes": 2649,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3059748411178589,
            "loss": 2.8238237351179123,
            "approx_kl": 0.003876123228110373,
            "wall_seconds": 2133.0938218999945
          },
          {
            "update": 261,
            "environment_steps": 1069056,
            "curriculum": 3,
            "completed_episodes": 2654,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2530001401901245,
            "loss": 9.262580394744873,
            "approx_kl": 0.0035672293015522882,
            "wall_seconds": 2138.955373899982
          },
          {
            "update": 262,
            "environment_steps": 1073152,
            "curriculum": 3,
            "completed_episodes": 2660,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.26449450850486755,
            "loss": 6.9586921483278275,
            "approx_kl": 0.003402614329388598,
            "wall_seconds": 2144.798768699984
          },
          {
            "update": 263,
            "environment_steps": 1077248,
            "curriculum": 3,
            "completed_episodes": 2664,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2647371292114258,
            "loss": 3.6922281980514526,
            "approx_kl": 0.002859330448700348,
            "wall_seconds": 2150.5388147999765
          },
          {
            "update": 264,
            "environment_steps": 1081344,
            "curriculum": 3,
            "completed_episodes": 2667,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.32312995195388794,
            "loss": 3.3038134276866913,
            "approx_kl": 0.007318280597246485,
            "wall_seconds": 2156.2565552000015
          },
          {
            "update": 265,
            "environment_steps": 1085440,
            "curriculum": 3,
            "completed_episodes": 2673,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.2912791073322296,
            "loss": 6.439081490039825,
            "approx_kl": 0.005027050421631429,
            "wall_seconds": 2162.1166416999768
          },
          {
            "update": 266,
            "environment_steps": 1089536,
            "curriculum": 3,
            "completed_episodes": 2678,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.27120301127433777,
            "loss": 5.107986479997635,
            "approx_kl": 0.004477641403354937,
            "wall_seconds": 2167.915629700001
          },
          {
            "update": 267,
            "environment_steps": 1093632,
            "curriculum": 3,
            "completed_episodes": 2681,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3071795105934143,
            "loss": 3.086769998073578,
            "approx_kl": 0.0038757946786063258,
            "wall_seconds": 2173.7160012999957
          },
          {
            "update": 268,
            "environment_steps": 1097728,
            "curriculum": 3,
            "completed_episodes": 2684,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2701556086540222,
            "loss": 4.99054254591465,
            "approx_kl": 0.0032607918401481584,
            "wall_seconds": 2179.5068324999884
          },
          {
            "update": 269,
            "environment_steps": 1101824,
            "curriculum": 3,
            "completed_episodes": 2685,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.2956492304801941,
            "loss": 2.3471526503562927,
            "approx_kl": 0.002152745695639169,
            "wall_seconds": 2185.2466914999823
          },
          {
            "update": 270,
            "environment_steps": 1105920,
            "curriculum": 3,
            "completed_episodes": 2688,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.310150682926178,
            "loss": 5.142621338367462,
            "approx_kl": 0.004372101757326163,
            "wall_seconds": 2189.1757666999765
          },
          {
            "update": 271,
            "environment_steps": 1110016,
            "curriculum": 3,
            "completed_episodes": 2697,
            "new_episodes": 9,
            "new_successes": 2,
            "mean_step_reward": -0.26802051067352295,
            "loss": 10.615708440542221,
            "approx_kl": 0.0036304868626757525,
            "wall_seconds": 2191.907460699993
          },
          {
            "update": 272,
            "environment_steps": 1114112,
            "curriculum": 3,
            "completed_episodes": 2703,
            "new_episodes": 6,
            "new_successes": 3,
            "mean_step_reward": -0.2729629874229431,
            "loss": 6.48836824297905,
            "approx_kl": 0.0046551366249332204,
            "wall_seconds": 2194.7092470999924
          },
          {
            "update": 273,
            "environment_steps": 1118208,
            "curriculum": 3,
            "completed_episodes": 2706,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2825075685977936,
            "loss": 5.225061044096947,
            "approx_kl": 0.004466555586986942,
            "wall_seconds": 2197.4412201999803
          },
          {
            "update": 274,
            "environment_steps": 1122304,
            "curriculum": 3,
            "completed_episodes": 2711,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.27667707204818726,
            "loss": 5.303915545344353,
            "approx_kl": 0.00388516188468202,
            "wall_seconds": 2200.1662672999955
          },
          {
            "update": 275,
            "environment_steps": 1126400,
            "curriculum": 3,
            "completed_episodes": 2716,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2788015604019165,
            "loss": 6.000712335109711,
            "approx_kl": 0.005178444633202162,
            "wall_seconds": 2202.9042386999936
          },
          {
            "update": 276,
            "environment_steps": 1130496,
            "curriculum": 3,
            "completed_episodes": 2719,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2966284155845642,
            "loss": 3.7928138822317123,
            "approx_kl": 0.0028538271762954537,
            "wall_seconds": 2205.6304492999916
          },
          {
            "update": 277,
            "environment_steps": 1134592,
            "curriculum": 3,
            "completed_episodes": 2723,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3097727298736572,
            "loss": 4.861142054200172,
            "approx_kl": 0.004964149517036276,
            "wall_seconds": 2208.3500138999952
          },
          {
            "update": 278,
            "environment_steps": 1138688,
            "curriculum": 3,
            "completed_episodes": 2725,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2980026602745056,
            "loss": 2.749582067131996,
            "approx_kl": 0.002741044525464531,
            "wall_seconds": 2211.0651233999815
          },
          {
            "update": 279,
            "environment_steps": 1142784,
            "curriculum": 3,
            "completed_episodes": 2728,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.3058227598667145,
            "loss": 4.35947185754776,
            "approx_kl": 0.003926415109162917,
            "wall_seconds": 2213.8016698999854
          },
          {
            "update": 280,
            "environment_steps": 1146880,
            "curriculum": 3,
            "completed_episodes": 2732,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2364797294139862,
            "loss": 6.011541321873665,
            "approx_kl": 0.003220207385311369,
            "wall_seconds": 2217.2123202999937
          },
          {
            "update": 281,
            "environment_steps": 1150976,
            "curriculum": 3,
            "completed_episodes": 2741,
            "new_episodes": 9,
            "new_successes": 2,
            "mean_step_reward": -0.2491731196641922,
            "loss": 8.605876952409744,
            "approx_kl": 0.003087687848164933,
            "wall_seconds": 2223.3334577999776
          },
          {
            "update": 282,
            "environment_steps": 1155072,
            "curriculum": 3,
            "completed_episodes": 2745,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2649747431278229,
            "loss": 5.365813419222832,
            "approx_kl": 0.002210738279245561,
            "wall_seconds": 2229.456768999982
          },
          {
            "update": 283,
            "environment_steps": 1159168,
            "curriculum": 3,
            "completed_episodes": 2748,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2894345223903656,
            "loss": 3.3461438044905663,
            "approx_kl": 0.004283056783606298,
            "wall_seconds": 2235.6411000999797
          },
          {
            "update": 284,
            "environment_steps": 1163264,
            "curriculum": 3,
            "completed_episodes": 2754,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2647362947463989,
            "loss": 7.773234844207764,
            "approx_kl": 0.004810347720194841,
            "wall_seconds": 2241.863418599998
          },
          {
            "update": 285,
            "environment_steps": 1167360,
            "curriculum": 3,
            "completed_episodes": 2761,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.2511988878250122,
            "loss": 8.733060777187347,
            "approx_kl": 0.002962619731988525,
            "wall_seconds": 2248.123066
          },
          {
            "update": 286,
            "environment_steps": 1171456,
            "curriculum": 3,
            "completed_episodes": 2764,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2373206913471222,
            "loss": 5.478142529726028,
            "approx_kl": 0.00584063796850387,
            "wall_seconds": 2254.3574611999793
          },
          {
            "update": 287,
            "environment_steps": 1175552,
            "curriculum": 3,
            "completed_episodes": 2768,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.23827023804187775,
            "loss": 8.24132776260376,
            "approx_kl": 0.0030437907262239605,
            "wall_seconds": 2260.601266499987
          },
          {
            "update": 288,
            "environment_steps": 1179648,
            "curriculum": 3,
            "completed_episodes": 2771,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.25281089544296265,
            "loss": 4.009245812892914,
            "approx_kl": 0.004641474632080644,
            "wall_seconds": 2264.8540458999923
          },
          {
            "update": 289,
            "environment_steps": 1183744,
            "curriculum": 3,
            "completed_episodes": 2776,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.31420210003852844,
            "loss": 7.493096590042114,
            "approx_kl": 0.0036503409573924728,
            "wall_seconds": 2267.5767027999973
          },
          {
            "update": 290,
            "environment_steps": 1187840,
            "curriculum": 3,
            "completed_episodes": 2783,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.32516348361968994,
            "loss": 7.070609122514725,
            "approx_kl": 0.0030289961978269275,
            "wall_seconds": 2270.297441200004
          },
          {
            "update": 291,
            "environment_steps": 1191936,
            "curriculum": 3,
            "completed_episodes": 2787,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2738214135169983,
            "loss": 5.311960011720657,
            "approx_kl": 0.004779216691531474,
            "wall_seconds": 2273.0482976999774
          },
          {
            "update": 292,
            "environment_steps": 1196032,
            "curriculum": 3,
            "completed_episodes": 2788,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.28380298614501953,
            "loss": 3.640447363257408,
            "approx_kl": 0.004054283319419483,
            "wall_seconds": 2275.7675052999984
          },
          {
            "update": 293,
            "environment_steps": 1200128,
            "curriculum": 3,
            "completed_episodes": 2791,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3161158859729767,
            "loss": 3.9618175476789474,
            "approx_kl": 0.0035975279643025715,
            "wall_seconds": 2278.5046986999805
          },
          {
            "update": 294,
            "environment_steps": 1204224,
            "curriculum": 3,
            "completed_episodes": 2796,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2720102071762085,
            "loss": 6.612877130508423,
            "approx_kl": 0.005528050711291144,
            "wall_seconds": 2281.2635500000033
          },
          {
            "update": 295,
            "environment_steps": 1208320,
            "curriculum": 3,
            "completed_episodes": 2799,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2876920700073242,
            "loss": 5.77826189994812,
            "approx_kl": 0.001280608517845394,
            "wall_seconds": 2283.994307599991
          },
          {
            "update": 296,
            "environment_steps": 1212416,
            "curriculum": 3,
            "completed_episodes": 2803,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.279789000749588,
            "loss": 6.483102306723595,
            "approx_kl": 0.004006373153970344,
            "wall_seconds": 2286.7208216
          },
          {
            "update": 297,
            "environment_steps": 1216512,
            "curriculum": 3,
            "completed_episodes": 2806,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2633821964263916,
            "loss": 3.8320243656635284,
            "approx_kl": 0.004983392940630438,
            "wall_seconds": 2289.4528938999865
          },
          {
            "update": 298,
            "environment_steps": 1220608,
            "curriculum": 3,
            "completed_episodes": 2811,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.26134079694747925,
            "loss": 9.271952718496323,
            "approx_kl": 0.002792546511045657,
            "wall_seconds": 2292.1986815999844
          },
          {
            "update": 299,
            "environment_steps": 1224704,
            "curriculum": 3,
            "completed_episodes": 2819,
            "new_episodes": 8,
            "new_successes": 1,
            "mean_step_reward": -0.2887725234031677,
            "loss": 9.782660454511642,
            "approx_kl": 0.004785749686561758,
            "wall_seconds": 2294.943871800002
          },
          {
            "update": 300,
            "environment_steps": 1228800,
            "curriculum": 3,
            "completed_episodes": 2825,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.2834417223930359,
            "loss": 6.024886518716812,
            "approx_kl": 0.0035190931121178437,
            "wall_seconds": 2297.6915310999902
          },
          {
            "update": 301,
            "environment_steps": 1232896,
            "curriculum": 3,
            "completed_episodes": 2827,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.29472213983535767,
            "loss": 2.9941226541996,
            "approx_kl": 0.006894230016769143,
            "wall_seconds": 2300.416358699993
          },
          {
            "update": 302,
            "environment_steps": 1236992,
            "curriculum": 3,
            "completed_episodes": 2828,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3102756142616272,
            "loss": 2.050884425640106,
            "approx_kl": 0.0055548973723489325,
            "wall_seconds": 2303.1333883999905
          },
          {
            "update": 303,
            "environment_steps": 1241088,
            "curriculum": 3,
            "completed_episodes": 2833,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.32639193534851074,
            "loss": 3.9400928020477295,
            "approx_kl": 0.003560079105227487,
            "wall_seconds": 2305.927035200002
          },
          {
            "update": 304,
            "environment_steps": 1245184,
            "curriculum": 3,
            "completed_episodes": 2838,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.26250723004341125,
            "loss": 7.365430578589439,
            "approx_kl": 0.003027835817192681,
            "wall_seconds": 2308.6672549999785
          },
          {
            "update": 305,
            "environment_steps": 1249280,
            "curriculum": 3,
            "completed_episodes": 2841,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.3235512375831604,
            "loss": 2.890991598367691,
            "approx_kl": 0.005123694267240353,
            "wall_seconds": 2311.373149399995
          },
          {
            "update": 306,
            "environment_steps": 1253376,
            "curriculum": 3,
            "completed_episodes": 2843,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30734455585479736,
            "loss": 2.8681088611483574,
            "approx_kl": 0.0031950995689840056,
            "wall_seconds": 2314.085898199992
          },
          {
            "update": 307,
            "environment_steps": 1257472,
            "curriculum": 3,
            "completed_episodes": 2847,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.29948318004608154,
            "loss": 3.707710936665535,
            "approx_kl": 0.005867999705515103,
            "wall_seconds": 2316.8172578999947
          },
          {
            "update": 308,
            "environment_steps": 1261568,
            "curriculum": 3,
            "completed_episodes": 2854,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.28563451766967773,
            "loss": 8.244011104106903,
            "approx_kl": 0.004656054552469868,
            "wall_seconds": 2319.5443142000004
          },
          {
            "update": 309,
            "environment_steps": 1265664,
            "curriculum": 3,
            "completed_episodes": 2862,
            "new_episodes": 8,
            "new_successes": 2,
            "mean_step_reward": -0.277324378490448,
            "loss": 8.225003331899643,
            "approx_kl": 0.0036710079657495953,
            "wall_seconds": 2322.2742569999828
          },
          {
            "update": 310,
            "environment_steps": 1269760,
            "curriculum": 3,
            "completed_episodes": 2866,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2356397956609726,
            "loss": 6.752224296331406,
            "approx_kl": 0.0015025961365608964,
            "wall_seconds": 2325.0324040999985
          },
          {
            "update": 311,
            "environment_steps": 1273856,
            "curriculum": 3,
            "completed_episodes": 2868,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.29187655448913574,
            "loss": 3.4063311517238617,
            "approx_kl": 0.0035899817776225973,
            "wall_seconds": 2328.8304363999923
          },
          {
            "update": 312,
            "environment_steps": 1277952,
            "curriculum": 3,
            "completed_episodes": 2871,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.26655155420303345,
            "loss": 8.56812670826912,
            "approx_kl": 0.0028700468319584616,
            "wall_seconds": 2334.792018899985
          },
          {
            "update": 313,
            "environment_steps": 1282048,
            "curriculum": 3,
            "completed_episodes": 2876,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2997456192970276,
            "loss": 5.449143886566162,
            "approx_kl": 0.0032128380189533345,
            "wall_seconds": 2340.974102099979
          },
          {
            "update": 314,
            "environment_steps": 1286144,
            "curriculum": 3,
            "completed_episodes": 2879,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2962211072444916,
            "loss": 5.208850592374802,
            "approx_kl": 0.005088896014058264,
            "wall_seconds": 2347.2179037999886
          },
          {
            "update": 315,
            "environment_steps": 1290240,
            "curriculum": 3,
            "completed_episodes": 2882,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2801344394683838,
            "loss": 7.570496320724487,
            "approx_kl": 0.0034386093611828983,
            "wall_seconds": 2353.433598100004
          },
          {
            "update": 316,
            "environment_steps": 1294336,
            "curriculum": 3,
            "completed_episodes": 2883,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3172714412212372,
            "loss": 1.6640928164124489,
            "approx_kl": 0.0030458732035185676,
            "wall_seconds": 2359.5702896000003
          },
          {
            "update": 317,
            "environment_steps": 1298432,
            "curriculum": 3,
            "completed_episodes": 2887,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.31617748737335205,
            "loss": 4.713567063212395,
            "approx_kl": 0.005103590985527262,
            "wall_seconds": 2365.728428100003
          },
          {
            "update": 318,
            "environment_steps": 1302528,
            "curriculum": 3,
            "completed_episodes": 2896,
            "new_episodes": 9,
            "new_successes": 1,
            "mean_step_reward": -0.29019394516944885,
            "loss": 7.949359029531479,
            "approx_kl": 0.0031255529502232093,
            "wall_seconds": 2371.9459041999944
          },
          {
            "update": 319,
            "environment_steps": 1306624,
            "curriculum": 3,
            "completed_episodes": 2901,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.300531804561615,
            "loss": 4.598897710442543,
            "approx_kl": 0.0037285929392965045,
            "wall_seconds": 2376.2048100999964
          },
          {
            "update": 320,
            "environment_steps": 1310720,
            "curriculum": 3,
            "completed_episodes": 2903,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.27760857343673706,
            "loss": 4.286983087658882,
            "approx_kl": 0.006191418964590412,
            "wall_seconds": 2378.9329714999767
          },
          {
            "update": 321,
            "environment_steps": 1314816,
            "curriculum": 3,
            "completed_episodes": 2906,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.30005085468292236,
            "loss": 3.8730476796627045,
            "approx_kl": 0.004510423426836496,
            "wall_seconds": 2381.655297899997
          },
          {
            "update": 322,
            "environment_steps": 1318912,
            "curriculum": 3,
            "completed_episodes": 2912,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.30802103877067566,
            "loss": 7.38864403963089,
            "approx_kl": 0.003528994846419664,
            "wall_seconds": 2384.398982599989
          },
          {
            "update": 323,
            "environment_steps": 1323008,
            "curriculum": 3,
            "completed_episodes": 2915,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2575996518135071,
            "loss": 4.033470526337624,
            "approx_kl": 0.0031372756529890466,
            "wall_seconds": 2387.124620799994
          },
          {
            "update": 324,
            "environment_steps": 1327104,
            "curriculum": 3,
            "completed_episodes": 2919,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2874126434326172,
            "loss": 3.574103981256485,
            "approx_kl": 0.006155129081889754,
            "wall_seconds": 2389.8547590999806
          },
          {
            "update": 325,
            "environment_steps": 1331200,
            "curriculum": 3,
            "completed_episodes": 2921,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.3014928102493286,
            "loss": 1.8379900865256786,
            "approx_kl": 0.004375532153062522,
            "wall_seconds": 2392.5663867999974
          },
          {
            "update": 326,
            "environment_steps": 1335296,
            "curriculum": 3,
            "completed_episodes": 2922,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3024781346321106,
            "loss": 1.535546213388443,
            "approx_kl": 0.007221445252071135,
            "wall_seconds": 2395.3862361999927
          },
          {
            "update": 327,
            "environment_steps": 1339392,
            "curriculum": 3,
            "completed_episodes": 2932,
            "new_episodes": 10,
            "new_successes": 2,
            "mean_step_reward": -0.30809083580970764,
            "loss": 11.336342126131058,
            "approx_kl": 0.0025309103839390446,
            "wall_seconds": 2398.151337499992
          },
          {
            "update": 328,
            "environment_steps": 1343488,
            "curriculum": 3,
            "completed_episodes": 2938,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.22534069418907166,
            "loss": 13.116774678230286,
            "approx_kl": 0.00330562609815388,
            "wall_seconds": 2400.887806899991
          },
          {
            "update": 329,
            "environment_steps": 1347584,
            "curriculum": 3,
            "completed_episodes": 2943,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.2644882798194885,
            "loss": 8.069284111261368,
            "approx_kl": 0.0005913478780712467,
            "wall_seconds": 2403.633300299989
          },
          {
            "update": 330,
            "environment_steps": 1351680,
            "curriculum": 3,
            "completed_episodes": 2947,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2958323657512665,
            "loss": 3.2096563279628754,
            "approx_kl": 0.004953071529598674,
            "wall_seconds": 2406.356480599992
          },
          {
            "update": 331,
            "environment_steps": 1355776,
            "curriculum": 3,
            "completed_episodes": 2952,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2950354218482971,
            "loss": 5.456581473350525,
            "approx_kl": 0.0033182353581651114,
            "wall_seconds": 2409.084636499989
          },
          {
            "update": 332,
            "environment_steps": 1359872,
            "curriculum": 3,
            "completed_episodes": 2953,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.2787898778915405,
            "loss": 2.378898024559021,
            "approx_kl": 0.0038397589341911953,
            "wall_seconds": 2411.8081174999825
          },
          {
            "update": 333,
            "environment_steps": 1363968,
            "curriculum": 3,
            "completed_episodes": 2958,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2928676903247833,
            "loss": 6.021656230092049,
            "approx_kl": 0.004379611204058165,
            "wall_seconds": 2414.5491025999945
          },
          {
            "update": 334,
            "environment_steps": 1368064,
            "curriculum": 3,
            "completed_episodes": 2960,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.31007641553878784,
            "loss": 2.572659842669964,
            "approx_kl": 0.005805545170005644,
            "wall_seconds": 2417.254656999983
          },
          {
            "update": 335,
            "environment_steps": 1372160,
            "curriculum": 3,
            "completed_episodes": 2962,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.3019339442253113,
            "loss": 1.2255809344351292,
            "approx_kl": 0.0055985387589316815,
            "wall_seconds": 2419.981566100003
          },
          {
            "update": 336,
            "environment_steps": 1376256,
            "curriculum": 3,
            "completed_episodes": 2967,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.32586848735809326,
            "loss": 5.159300059080124,
            "approx_kl": 0.004803766598342918,
            "wall_seconds": 2422.6724616999854
          },
          {
            "update": 337,
            "environment_steps": 1380352,
            "curriculum": 3,
            "completed_episodes": 2976,
            "new_episodes": 9,
            "new_successes": 2,
            "mean_step_reward": -0.28807055950164795,
            "loss": 11.43104961514473,
            "approx_kl": 0.0022668753917969298,
            "wall_seconds": 2425.400078299979
          },
          {
            "update": 338,
            "environment_steps": 1384448,
            "curriculum": 3,
            "completed_episodes": 2980,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2903457283973694,
            "loss": 5.193300738930702,
            "approx_kl": 0.004705708332039649,
            "wall_seconds": 2428.1286967999768
          },
          {
            "update": 339,
            "environment_steps": 1388544,
            "curriculum": 3,
            "completed_episodes": 2984,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.286095529794693,
            "loss": 8.51168417930603,
            "approx_kl": 0.0039075720342225395,
            "wall_seconds": 2430.8490131
          },
          {
            "update": 340,
            "environment_steps": 1392640,
            "curriculum": 3,
            "completed_episodes": 2987,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.28023236989974976,
            "loss": 3.665483057498932,
            "approx_kl": 0.005534394615096971,
            "wall_seconds": 2433.577342000004
          },
          {
            "update": 341,
            "environment_steps": 1396736,
            "curriculum": 3,
            "completed_episodes": 2992,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.26787102222442627,
            "loss": 6.728060632944107,
            "approx_kl": 0.0034041998369502835,
            "wall_seconds": 2436.3484085999953
          },
          {
            "update": 342,
            "environment_steps": 1400832,
            "curriculum": 3,
            "completed_episodes": 2996,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.31647396087646484,
            "loss": 4.130748420953751,
            "approx_kl": 0.0069320464790507685,
            "wall_seconds": 2439.064604799991
          },
          {
            "update": 343,
            "environment_steps": 1404928,
            "curriculum": 3,
            "completed_episodes": 2998,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.299199640750885,
            "loss": 4.032894283533096,
            "approx_kl": 0.004220325372443767,
            "wall_seconds": 2441.776720599999
          },
          {
            "update": 344,
            "environment_steps": 1409024,
            "curriculum": 3,
            "completed_episodes": 2999,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.30817538499832153,
            "loss": 2.342628188431263,
            "approx_kl": 0.006444398957683006,
            "wall_seconds": 2444.4892677000025
          },
          {
            "update": 345,
            "environment_steps": 1413120,
            "curriculum": 3,
            "completed_episodes": 3001,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.3096785545349121,
            "loss": 1.7426541075110435,
            "approx_kl": 0.005164151705685072,
            "wall_seconds": 2447.1967836999975
          },
          {
            "update": 346,
            "environment_steps": 1417216,
            "curriculum": 3,
            "completed_episodes": 3011,
            "new_episodes": 10,
            "new_successes": 1,
            "mean_step_reward": -0.28824102878570557,
            "loss": 12.209035515785217,
            "approx_kl": 0.003448690225923201,
            "wall_seconds": 2449.9239060999826
          },
          {
            "update": 347,
            "environment_steps": 1421312,
            "curriculum": 3,
            "completed_episodes": 3020,
            "new_episodes": 9,
            "new_successes": 3,
            "mean_step_reward": -0.24483747780323029,
            "loss": 6.680842190980911,
            "approx_kl": 0.004636561207007617,
            "wall_seconds": 2452.6524928
          },
          {
            "update": 348,
            "environment_steps": 1425408,
            "curriculum": 3,
            "completed_episodes": 3024,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2598104774951935,
            "loss": 5.562630355358124,
            "approx_kl": 0.004371339047793299,
            "wall_seconds": 2455.384559999977
          },
          {
            "update": 349,
            "environment_steps": 1429504,
            "curriculum": 3,
            "completed_episodes": 3025,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.2918170094490051,
            "loss": 4.038191549479961,
            "approx_kl": 0.004201040988846216,
            "wall_seconds": 2458.148784399993
          },
          {
            "update": 350,
            "environment_steps": 1433600,
            "curriculum": 3,
            "completed_episodes": 3032,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2953053116798401,
            "loss": 8.966781824827194,
            "approx_kl": 0.00321337697096169,
            "wall_seconds": 2460.9186777999857
          },
          {
            "update": 351,
            "environment_steps": 1437696,
            "curriculum": 3,
            "completed_episodes": 3035,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.26411762833595276,
            "loss": 2.7123395577073097,
            "approx_kl": 0.005203638345847139,
            "wall_seconds": 2463.651299099991
          },
          {
            "update": 352,
            "environment_steps": 1441792,
            "curriculum": 3,
            "completed_episodes": 3037,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3028072118759155,
            "loss": 3.2474510818719864,
            "approx_kl": 0.005204230728850234,
            "wall_seconds": 2466.37837999998
          },
          {
            "update": 353,
            "environment_steps": 1445888,
            "curriculum": 3,
            "completed_episodes": 3038,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.3127153813838959,
            "loss": 3.005432680249214,
            "approx_kl": 0.007012817837676266,
            "wall_seconds": 2469.0996907999797
          },
          {
            "update": 354,
            "environment_steps": 1449984,
            "curriculum": 3,
            "completed_episodes": 3040,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.32346051931381226,
            "loss": 2.322038196027279,
            "approx_kl": 0.0054358970701287035,
            "wall_seconds": 2471.8145103999996
          },
          {
            "update": 355,
            "environment_steps": 1454080,
            "curriculum": 3,
            "completed_episodes": 3044,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3299593925476074,
            "loss": 4.704375043511391,
            "approx_kl": 0.00560542824678123,
            "wall_seconds": 2474.5282289999886
          },
          {
            "update": 356,
            "environment_steps": 1458176,
            "curriculum": 3,
            "completed_episodes": 3051,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.26492100954055786,
            "loss": 7.588563114404678,
            "approx_kl": 0.007066887144901557,
            "wall_seconds": 2477.2490239999897
          },
          {
            "update": 357,
            "environment_steps": 1462272,
            "curriculum": 3,
            "completed_episodes": 3056,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.22752425074577332,
            "loss": 8.459702968597412,
            "approx_kl": 0.0038347105946741067,
            "wall_seconds": 2479.997536699986
          },
          {
            "update": 358,
            "environment_steps": 1466368,
            "curriculum": 3,
            "completed_episodes": 3061,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2938789129257202,
            "loss": 6.012048989534378,
            "approx_kl": 0.004906673937512096,
            "wall_seconds": 2482.741748100001
          },
          {
            "update": 359,
            "environment_steps": 1470464,
            "curriculum": 3,
            "completed_episodes": 3064,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2923208773136139,
            "loss": 4.392767012119293,
            "approx_kl": 0.004679048994148616,
            "wall_seconds": 2485.4661298999854
          },
          {
            "update": 360,
            "environment_steps": 1474560,
            "curriculum": 3,
            "completed_episodes": 3068,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3020176887512207,
            "loss": 5.601586490869522,
            "approx_kl": 0.0049224813883483876,
            "wall_seconds": 2488.183263599989
          },
          {
            "update": 361,
            "environment_steps": 1478656,
            "curriculum": 3,
            "completed_episodes": 3071,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2968143820762634,
            "loss": 4.078884065151215,
            "approx_kl": 0.004321289718063781,
            "wall_seconds": 2490.928709699976
          },
          {
            "update": 362,
            "environment_steps": 1482752,
            "curriculum": 3,
            "completed_episodes": 3073,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3007334768772125,
            "loss": 2.5737295672297478,
            "approx_kl": 0.005129046028741868,
            "wall_seconds": 2493.6446182999935
          },
          {
            "update": 363,
            "environment_steps": 1486848,
            "curriculum": 3,
            "completed_episodes": 3075,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.29624447226524353,
            "loss": 2.985018417239189,
            "approx_kl": 0.008866544198099291,
            "wall_seconds": 2496.3746452999767
          },
          {
            "update": 364,
            "environment_steps": 1490944,
            "curriculum": 3,
            "completed_episodes": 3077,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30079978704452515,
            "loss": 2.6688937172293663,
            "approx_kl": 0.008940604624513071,
            "wall_seconds": 2499.0985194999957
          },
          {
            "update": 365,
            "environment_steps": 1495040,
            "curriculum": 3,
            "completed_episodes": 3083,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.2675521969795227,
            "loss": 6.2372108697891235,
            "approx_kl": 0.011138714908156544,
            "wall_seconds": 2501.8280775999883
          },
          {
            "update": 366,
            "environment_steps": 1499136,
            "curriculum": 3,
            "completed_episodes": 3091,
            "new_episodes": 8,
            "new_successes": 1,
            "mean_step_reward": -0.21686169505119324,
            "loss": 9.884591281414032,
            "approx_kl": 0.0037961678935971577,
            "wall_seconds": 2504.582063399983
          },
          {
            "update": 367,
            "environment_steps": 1503232,
            "curriculum": 3,
            "completed_episodes": 3097,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.26609736680984497,
            "loss": 5.341448187828064,
            "approx_kl": 0.006809276470448822,
            "wall_seconds": 2507.31421349998
          },
          {
            "update": 368,
            "environment_steps": 1507328,
            "curriculum": 3,
            "completed_episodes": 3097,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.3104037344455719,
            "loss": 0.7220536842942238,
            "approx_kl": 0.0034345733183727134,
            "wall_seconds": 2510.046875400003
          },
          {
            "update": 369,
            "environment_steps": 1511424,
            "curriculum": 3,
            "completed_episodes": 3103,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.31263941526412964,
            "loss": 6.303300470113754,
            "approx_kl": 0.0024982875838759355,
            "wall_seconds": 2512.8248598999926
          },
          {
            "update": 370,
            "environment_steps": 1515520,
            "curriculum": 3,
            "completed_episodes": 3105,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.28242599964141846,
            "loss": 2.9172296077013016,
            "approx_kl": 0.006009360800817376,
            "wall_seconds": 2515.553063600004
          },
          {
            "update": 371,
            "environment_steps": 1519616,
            "curriculum": 3,
            "completed_episodes": 3107,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3267850875854492,
            "loss": 3.442000538110733,
            "approx_kl": 0.005017487914301455,
            "wall_seconds": 2518.2975353999936
          },
          {
            "update": 372,
            "environment_steps": 1523712,
            "curriculum": 3,
            "completed_episodes": 3108,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3020222783088684,
            "loss": 1.4817034639418125,
            "approx_kl": 0.006463432455348084,
            "wall_seconds": 2521.029946900002
          },
          {
            "update": 373,
            "environment_steps": 1527808,
            "curriculum": 3,
            "completed_episodes": 3109,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3074021339416504,
            "loss": 1.8778501600027084,
            "approx_kl": 0.01257075612738845,
            "wall_seconds": 2523.745439899998
          },
          {
            "update": 374,
            "environment_steps": 1531904,
            "curriculum": 3,
            "completed_episodes": 3115,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.30339109897613525,
            "loss": 8.17380827665329,
            "approx_kl": 0.00630188808645471,
            "wall_seconds": 2526.4776789000025
          },
          {
            "update": 375,
            "environment_steps": 1536000,
            "curriculum": 3,
            "completed_episodes": 3125,
            "new_episodes": 10,
            "new_successes": 1,
            "mean_step_reward": -0.2518877685070038,
            "loss": 12.106874108314514,
            "approx_kl": 0.003595925143599743,
            "wall_seconds": 2529.2328884000017
          },
          {
            "update": 376,
            "environment_steps": 1540096,
            "curriculum": 3,
            "completed_episodes": 3133,
            "new_episodes": 8,
            "new_successes": 4,
            "mean_step_reward": -0.21922439336776733,
            "loss": 12.490780293941498,
            "approx_kl": 0.002968840279208962,
            "wall_seconds": 2531.975055399991
          },
          {
            "update": 377,
            "environment_steps": 1544192,
            "curriculum": 3,
            "completed_episodes": 3135,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2505062520503998,
            "loss": 5.682602971792221,
            "approx_kl": 0.003057197402085876,
            "wall_seconds": 2534.71116169999
          },
          {
            "update": 378,
            "environment_steps": 1548288,
            "curriculum": 3,
            "completed_episodes": 3142,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2532990872859955,
            "loss": 7.24272495508194,
            "approx_kl": 0.005438021769805346,
            "wall_seconds": 2537.445839199994
          },
          {
            "update": 379,
            "environment_steps": 1552384,
            "curriculum": 3,
            "completed_episodes": 3143,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.23117101192474365,
            "loss": 5.221507370471954,
            "approx_kl": 0.005593622110609431,
            "wall_seconds": 2540.201071299991
          },
          {
            "update": 380,
            "environment_steps": 1556480,
            "curriculum": 3,
            "completed_episodes": 3149,
            "new_episodes": 6,
            "new_successes": 3,
            "mean_step_reward": -0.23044095933437347,
            "loss": 7.556563764810562,
            "approx_kl": 0.002933671941718785,
            "wall_seconds": 2542.9334490000037
          },
          {
            "update": 381,
            "environment_steps": 1560576,
            "curriculum": 3,
            "completed_episodes": 3149,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.2895381450653076,
            "loss": 1.5651510879397392,
            "approx_kl": 0.003512458762997994,
            "wall_seconds": 2545.7042227999773
          },
          {
            "update": 382,
            "environment_steps": 1564672,
            "curriculum": 3,
            "completed_episodes": 3150,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.30387330055236816,
            "loss": 2.233720302581787,
            "approx_kl": 0.007597250340040773,
            "wall_seconds": 2548.413298999978
          },
          {
            "update": 383,
            "environment_steps": 1568768,
            "curriculum": 3,
            "completed_episodes": 3152,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30352818965911865,
            "loss": 2.8611471503973007,
            "approx_kl": 0.006207434769748943,
            "wall_seconds": 2551.1378122000024
          },
          {
            "update": 384,
            "environment_steps": 1572864,
            "curriculum": 3,
            "completed_episodes": 3158,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.3213402330875397,
            "loss": 6.082003116607666,
            "approx_kl": 0.004788819551322376,
            "wall_seconds": 2553.850711499981
          },
          {
            "update": 385,
            "environment_steps": 1576960,
            "curriculum": 3,
            "completed_episodes": 3166,
            "new_episodes": 8,
            "new_successes": 0,
            "mean_step_reward": -0.3047792315483093,
            "loss": 6.765110194683075,
            "approx_kl": 0.00447079372315784,
            "wall_seconds": 2556.5863101999857
          },
          {
            "update": 386,
            "environment_steps": 1581056,
            "curriculum": 3,
            "completed_episodes": 3172,
            "new_episodes": 6,
            "new_successes": 3,
            "mean_step_reward": -0.2739848792552948,
            "loss": 8.75722786784172,
            "approx_kl": 0.005421636738901725,
            "wall_seconds": 2559.3151408999984
          },
          {
            "update": 387,
            "environment_steps": 1585152,
            "curriculum": 3,
            "completed_episodes": 3177,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.30387118458747864,
            "loss": 5.453480571508408,
            "approx_kl": 0.002852820478437934,
            "wall_seconds": 2562.03699589998
          },
          {
            "update": 388,
            "environment_steps": 1589248,
            "curriculum": 3,
            "completed_episodes": 3179,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.28915518522262573,
            "loss": 5.392579957842827,
            "approx_kl": 0.006456637194787618,
            "wall_seconds": 2564.7739106999943
          },
          {
            "update": 389,
            "environment_steps": 1593344,
            "curriculum": 3,
            "completed_episodes": 3184,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.30446821451187134,
            "loss": 7.457904279232025,
            "approx_kl": 0.005134137027198449,
            "wall_seconds": 2567.4954136000015
          },
          {
            "update": 390,
            "environment_steps": 1597440,
            "curriculum": 3,
            "completed_episodes": 3187,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2826331555843353,
            "loss": 6.11517421901226,
            "approx_kl": 0.005941039406025084,
            "wall_seconds": 2570.233215699991
          },
          {
            "update": 391,
            "environment_steps": 1601536,
            "curriculum": 3,
            "completed_episodes": 3189,
            "new_episodes": 2,
            "new_successes": 2,
            "mean_step_reward": -0.299297571182251,
            "loss": 4.032199069857597,
            "approx_kl": 0.004191621617792407,
            "wall_seconds": 2572.956897099997
          },
          {
            "update": 392,
            "environment_steps": 1605632,
            "curriculum": 3,
            "completed_episodes": 3191,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3297846019268036,
            "loss": 1.8843553774058819,
            "approx_kl": 0.00474666282752878,
            "wall_seconds": 2575.661680099991
          },
          {
            "update": 393,
            "environment_steps": 1609728,
            "curriculum": 3,
            "completed_episodes": 3194,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.31009969115257263,
            "loss": 4.391514420509338,
            "approx_kl": 0.006220767958438955,
            "wall_seconds": 2578.38313609999
          },
          {
            "update": 394,
            "environment_steps": 1613824,
            "curriculum": 3,
            "completed_episodes": 3201,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2596295475959778,
            "loss": 7.509469002485275,
            "approx_kl": 0.006316843264357885,
            "wall_seconds": 2581.1117744999938
          },
          {
            "update": 395,
            "environment_steps": 1617920,
            "curriculum": 3,
            "completed_episodes": 3208,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.29014819860458374,
            "loss": 7.193902105093002,
            "approx_kl": 0.004520952959865099,
            "wall_seconds": 2583.835086199979
          },
          {
            "update": 396,
            "environment_steps": 1622016,
            "curriculum": 3,
            "completed_episodes": 3211,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.304259717464447,
            "loss": 3.8917502015829086,
            "approx_kl": 0.0031562620570184663,
            "wall_seconds": 2586.5508296999906
          },
          {
            "update": 397,
            "environment_steps": 1626112,
            "curriculum": 3,
            "completed_episodes": 3216,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.30721455812454224,
            "loss": 4.731085196137428,
            "approx_kl": 0.003419221658987226,
            "wall_seconds": 2589.2791977000015
          },
          {
            "update": 398,
            "environment_steps": 1630208,
            "curriculum": 3,
            "completed_episodes": 3221,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.28154993057250977,
            "loss": 5.08165530860424,
            "approx_kl": 0.0034188589379482437,
            "wall_seconds": 2592.0056639999966
          },
          {
            "update": 399,
            "environment_steps": 1634304,
            "curriculum": 3,
            "completed_episodes": 3227,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.30436575412750244,
            "loss": 6.290653213858604,
            "approx_kl": 0.0028236257421667688,
            "wall_seconds": 2594.7485667
          },
          {
            "update": 400,
            "environment_steps": 1638400,
            "curriculum": 3,
            "completed_episodes": 3227,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.3139779269695282,
            "loss": 0.944741066545248,
            "approx_kl": 0.008486302918754518,
            "wall_seconds": 2597.459178099991
          },
          {
            "update": 401,
            "environment_steps": 1642496,
            "curriculum": 3,
            "completed_episodes": 3229,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.32930630445480347,
            "loss": 1.9769533425569534,
            "approx_kl": 0.015386495604616357,
            "wall_seconds": 2600.194246799976
          },
          {
            "update": 402,
            "environment_steps": 1646592,
            "curriculum": 3,
            "completed_episodes": 3232,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3321002721786499,
            "loss": 3.200985938310623,
            "approx_kl": 0.009320581208157819,
            "wall_seconds": 2602.8910280999844
          },
          {
            "update": 403,
            "environment_steps": 1650688,
            "curriculum": 3,
            "completed_episodes": 3239,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.27536094188690186,
            "loss": 7.430703550577164,
            "approx_kl": 0.004613376626366517,
            "wall_seconds": 2605.6129789999977
          },
          {
            "update": 404,
            "environment_steps": 1654784,
            "curriculum": 3,
            "completed_episodes": 3246,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2843407988548279,
            "loss": 7.008743852376938,
            "approx_kl": 0.0036025634726684075,
            "wall_seconds": 2608.3490376999835
          },
          {
            "update": 405,
            "environment_steps": 1658880,
            "curriculum": 3,
            "completed_episodes": 3248,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.24952766299247742,
            "loss": 3.9388488978147507,
            "approx_kl": 0.0036667616295744665,
            "wall_seconds": 2611.1279462999955
          },
          {
            "update": 406,
            "environment_steps": 1662976,
            "curriculum": 3,
            "completed_episodes": 3252,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.31159430742263794,
            "loss": 5.693092942237854,
            "approx_kl": 0.0026072259570355527,
            "wall_seconds": 2613.850543699984
          },
          {
            "update": 407,
            "environment_steps": 1667072,
            "curriculum": 3,
            "completed_episodes": 3254,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2863702178001404,
            "loss": 4.06229355931282,
            "approx_kl": 0.0042022978705063,
            "wall_seconds": 2616.6086679
          },
          {
            "update": 408,
            "environment_steps": 1671168,
            "curriculum": 3,
            "completed_episodes": 3261,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.3165930509567261,
            "loss": 8.387454628944397,
            "approx_kl": 0.005805862718261778,
            "wall_seconds": 2619.3702805999783
          },
          {
            "update": 409,
            "environment_steps": 1675264,
            "curriculum": 3,
            "completed_episodes": 3264,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2932373285293579,
            "loss": 5.7018944919109344,
            "approx_kl": 0.00259382982403622,
            "wall_seconds": 2622.091069599992
          },
          {
            "update": 410,
            "environment_steps": 1679360,
            "curriculum": 3,
            "completed_episodes": 3266,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.32509979605674744,
            "loss": 2.5795478969812393,
            "approx_kl": 0.005655480614223052,
            "wall_seconds": 2624.8155540000007
          },
          {
            "update": 411,
            "environment_steps": 1683456,
            "curriculum": 3,
            "completed_episodes": 3267,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.32584160566329956,
            "loss": 1.4255251288414001,
            "approx_kl": 0.007580857403809205,
            "wall_seconds": 2627.5240000999765
          },
          {
            "update": 412,
            "environment_steps": 1687552,
            "curriculum": 3,
            "completed_episodes": 3271,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.32308897376060486,
            "loss": 5.687567964196205,
            "approx_kl": 0.008640482654300286,
            "wall_seconds": 2630.2236389000027
          },
          {
            "update": 413,
            "environment_steps": 1691648,
            "curriculum": 3,
            "completed_episodes": 3279,
            "new_episodes": 8,
            "new_successes": 2,
            "mean_step_reward": -0.29709476232528687,
            "loss": 11.496803224086761,
            "approx_kl": 0.0039816617681935895,
            "wall_seconds": 2632.9517166999867
          },
          {
            "update": 414,
            "environment_steps": 1695744,
            "curriculum": 3,
            "completed_episodes": 3284,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.28072589635849,
            "loss": 4.618954449892044,
            "approx_kl": 0.0035212391339882743,
            "wall_seconds": 2635.6744183999836
          },
          {
            "update": 415,
            "environment_steps": 1699840,
            "curriculum": 3,
            "completed_episodes": 3290,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.30197960138320923,
            "loss": 7.178732544183731,
            "approx_kl": 0.00220810803148197,
            "wall_seconds": 2638.4012806999963
          },
          {
            "update": 416,
            "environment_steps": 1703936,
            "curriculum": 3,
            "completed_episodes": 3291,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.30512237548828125,
            "loss": 1.845544658601284,
            "approx_kl": 0.00399237724559498,
            "wall_seconds": 2641.1192845999904
          },
          {
            "update": 417,
            "environment_steps": 1708032,
            "curriculum": 3,
            "completed_episodes": 3297,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.3104174733161926,
            "loss": 7.059171438217163,
            "approx_kl": 0.005031652748584747,
            "wall_seconds": 2643.828737599979
          },
          {
            "update": 418,
            "environment_steps": 1712128,
            "curriculum": 3,
            "completed_episodes": 3303,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.2679618000984192,
            "loss": 8.529945582151413,
            "approx_kl": 0.003704001646838151,
            "wall_seconds": 2646.562483899994
          },
          {
            "update": 419,
            "environment_steps": 1716224,
            "curriculum": 3,
            "completed_episodes": 3306,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2768841087818146,
            "loss": 4.044925093650818,
            "approx_kl": 0.005262184757157229,
            "wall_seconds": 2649.298770399997
          },
          {
            "update": 420,
            "environment_steps": 1720320,
            "curriculum": 3,
            "completed_episodes": 3308,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.29717400670051575,
            "loss": 2.4785172566771507,
            "approx_kl": 0.0038870732896612026,
            "wall_seconds": 2652.0207941999834
          },
          {
            "update": 421,
            "environment_steps": 1724416,
            "curriculum": 3,
            "completed_episodes": 3310,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3274826109409332,
            "loss": 3.1132406890392303,
            "approx_kl": 0.00503289014523034,
            "wall_seconds": 2654.784098199976
          },
          {
            "update": 422,
            "environment_steps": 1728512,
            "curriculum": 3,
            "completed_episodes": 3315,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.32809677720069885,
            "loss": 6.449863463640213,
            "approx_kl": 0.002908712391217705,
            "wall_seconds": 2657.502013899997
          },
          {
            "update": 423,
            "environment_steps": 1732608,
            "curriculum": 3,
            "completed_episodes": 3323,
            "new_episodes": 8,
            "new_successes": 2,
            "mean_step_reward": -0.2961876094341278,
            "loss": 9.584080040454865,
            "approx_kl": 0.0018428860530548263,
            "wall_seconds": 2660.23684279999
          },
          {
            "update": 424,
            "environment_steps": 1736704,
            "curriculum": 3,
            "completed_episodes": 3325,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.305279016494751,
            "loss": 1.1801149733364582,
            "approx_kl": 0.003083661966229556,
            "wall_seconds": 2662.969518400001
          },
          {
            "update": 425,
            "environment_steps": 1740800,
            "curriculum": 3,
            "completed_episodes": 3332,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.29780203104019165,
            "loss": 6.1527209877967834,
            "approx_kl": 0.006149302800622536,
            "wall_seconds": 2665.7482335999957
          },
          {
            "update": 426,
            "environment_steps": 1744896,
            "curriculum": 3,
            "completed_episodes": 3337,
            "new_episodes": 5,
            "new_successes": 3,
            "mean_step_reward": -0.2707735598087311,
            "loss": 11.099705636501312,
            "approx_kl": 0.004091693801456131,
            "wall_seconds": 2668.486303199985
          },
          {
            "update": 427,
            "environment_steps": 1748992,
            "curriculum": 3,
            "completed_episodes": 3343,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.31161361932754517,
            "loss": 6.467480540275574,
            "approx_kl": 0.0037802845909027383,
            "wall_seconds": 2671.2196822999977
          },
          {
            "update": 428,
            "environment_steps": 1753088,
            "curriculum": 3,
            "completed_episodes": 3345,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.31590190529823303,
            "loss": 2.542132295668125,
            "approx_kl": 0.003161024709697813,
            "wall_seconds": 2673.9406132999866
          },
          {
            "update": 429,
            "environment_steps": 1757184,
            "curriculum": 3,
            "completed_episodes": 3350,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.28652870655059814,
            "loss": 8.871611267328262,
            "approx_kl": 0.003076292388868751,
            "wall_seconds": 2676.6670153999876
          },
          {
            "update": 430,
            "environment_steps": 1761280,
            "curriculum": 3,
            "completed_episodes": 3352,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.29244551062583923,
            "loss": 5.505283638834953,
            "approx_kl": 0.008363036231457954,
            "wall_seconds": 2679.3850714
          },
          {
            "update": 431,
            "environment_steps": 1765376,
            "curriculum": 3,
            "completed_episodes": 3356,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.28616294264793396,
            "loss": 6.432640492916107,
            "approx_kl": 0.004119752589758718,
            "wall_seconds": 2682.1295365999977
          },
          {
            "update": 432,
            "environment_steps": 1769472,
            "curriculum": 3,
            "completed_episodes": 3362,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.29291093349456787,
            "loss": 7.11545193195343,
            "approx_kl": 0.004779491158842575,
            "wall_seconds": 2684.8506834
          },
          {
            "update": 433,
            "environment_steps": 1773568,
            "curriculum": 3,
            "completed_episodes": 3367,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2888548672199249,
            "loss": 6.208645075559616,
            "approx_kl": 0.0036917775505571626,
            "wall_seconds": 2687.575275799987
          },
          {
            "update": 434,
            "environment_steps": 1777664,
            "curriculum": 3,
            "completed_episodes": 3373,
            "new_episodes": 6,
            "new_successes": 4,
            "mean_step_reward": -0.2868354022502899,
            "loss": 7.507347583770752,
            "approx_kl": 0.0027800546704384033,
            "wall_seconds": 2690.302072099992
          },
          {
            "update": 435,
            "environment_steps": 1781760,
            "curriculum": 3,
            "completed_episodes": 3376,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.31425559520721436,
            "loss": 3.8846841156482697,
            "approx_kl": 0.0040788884507492185,
            "wall_seconds": 2693.0202027999912
          },
          {
            "update": 436,
            "environment_steps": 1785856,
            "curriculum": 3,
            "completed_episodes": 3379,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3368380069732666,
            "loss": 3.527628466486931,
            "approx_kl": 0.0034799360619217623,
            "wall_seconds": 2695.7096063999925
          },
          {
            "update": 437,
            "environment_steps": 1789952,
            "curriculum": 3,
            "completed_episodes": 3382,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3161033093929291,
            "loss": 3.042234845459461,
            "approx_kl": 0.006717264128383249,
            "wall_seconds": 2698.4285485
          },
          {
            "update": 438,
            "environment_steps": 1794048,
            "curriculum": 3,
            "completed_episodes": 3385,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.3138168454170227,
            "loss": 0.999979056417942,
            "approx_kl": 0.006850946538179414,
            "wall_seconds": 2701.1415870999917
          },
          {
            "update": 439,
            "environment_steps": 1798144,
            "curriculum": 3,
            "completed_episodes": 3390,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.3130679726600647,
            "loss": 6.072329327464104,
            "approx_kl": 0.0068501635832944885,
            "wall_seconds": 2703.9695629999915
          },
          {
            "update": 440,
            "environment_steps": 1802240,
            "curriculum": 3,
            "completed_episodes": 3392,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.30373337864875793,
            "loss": 2.056820787489414,
            "approx_kl": 0.004164559806667967,
            "wall_seconds": 2706.68905299998
          },
          {
            "update": 441,
            "environment_steps": 1806336,
            "curriculum": 3,
            "completed_episodes": 3397,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.305946409702301,
            "loss": 6.924012124538422,
            "approx_kl": 0.0043764128677139524,
            "wall_seconds": 2709.435814299999
          },
          {
            "update": 442,
            "environment_steps": 1810432,
            "curriculum": 3,
            "completed_episodes": 3403,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.29388564825057983,
            "loss": 6.5785738825798035,
            "approx_kl": 0.0058663078052632045,
            "wall_seconds": 2712.2229371999856
          },
          {
            "update": 443,
            "environment_steps": 1814528,
            "curriculum": 3,
            "completed_episodes": 3405,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3277486264705658,
            "loss": 2.5788792744278908,
            "approx_kl": 0.007527693705924321,
            "wall_seconds": 2714.982444299996
          },
          {
            "update": 444,
            "environment_steps": 1818624,
            "curriculum": 3,
            "completed_episodes": 3410,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2979814112186432,
            "loss": 6.392689287662506,
            "approx_kl": 0.0036381542631716,
            "wall_seconds": 2717.7045061999816
          },
          {
            "update": 445,
            "environment_steps": 1822720,
            "curriculum": 3,
            "completed_episodes": 3414,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.270682156085968,
            "loss": 4.05524043738842,
            "approx_kl": 0.003787846966588404,
            "wall_seconds": 2720.438722199993
          },
          {
            "update": 446,
            "environment_steps": 1826816,
            "curriculum": 3,
            "completed_episodes": 3418,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3027496933937073,
            "loss": 4.7098987102508545,
            "approx_kl": 0.0038592696691921446,
            "wall_seconds": 2723.1661454999994
          },
          {
            "update": 447,
            "environment_steps": 1830912,
            "curriculum": 3,
            "completed_episodes": 3420,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2735449969768524,
            "loss": 3.8717049807310104,
            "approx_kl": 0.005600919532298576,
            "wall_seconds": 2725.8906466999906
          },
          {
            "update": 448,
            "environment_steps": 1835008,
            "curriculum": 3,
            "completed_episodes": 3426,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.28272172808647156,
            "loss": 7.960175812244415,
            "approx_kl": 0.00412136070735869,
            "wall_seconds": 2728.611579999997
          },
          {
            "update": 449,
            "environment_steps": 1839104,
            "curriculum": 3,
            "completed_episodes": 3430,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2537039518356323,
            "loss": 10.14328619837761,
            "approx_kl": 0.0035984800160804298,
            "wall_seconds": 2731.3470270999824
          },
          {
            "update": 450,
            "environment_steps": 1843200,
            "curriculum": 3,
            "completed_episodes": 3437,
            "new_episodes": 7,
            "new_successes": 3,
            "mean_step_reward": -0.2958935499191284,
            "loss": 8.201210290193558,
            "approx_kl": 0.003032016131328419,
            "wall_seconds": 2734.0621594999975
          },
          {
            "update": 451,
            "environment_steps": 1847296,
            "curriculum": 3,
            "completed_episodes": 3443,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.30773478746414185,
            "loss": 6.20976048707962,
            "approx_kl": 0.004622541007847758,
            "wall_seconds": 2736.8285988999996
          },
          {
            "update": 452,
            "environment_steps": 1851392,
            "curriculum": 3,
            "completed_episodes": 3446,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2836810350418091,
            "loss": 3.2579739838838577,
            "approx_kl": 0.006646784364420455,
            "wall_seconds": 2739.567073299986
          },
          {
            "update": 453,
            "environment_steps": 1855488,
            "curriculum": 3,
            "completed_episodes": 3450,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.30487820506095886,
            "loss": 3.5602132380008698,
            "approx_kl": 0.007653877495613415,
            "wall_seconds": 2742.332595099986
          },
          {
            "update": 454,
            "environment_steps": 1859584,
            "curriculum": 3,
            "completed_episodes": 3454,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2859792709350586,
            "loss": 4.691430360078812,
            "approx_kl": 0.005364019161788747,
            "wall_seconds": 2745.085445899982
          },
          {
            "update": 455,
            "environment_steps": 1863680,
            "curriculum": 3,
            "completed_episodes": 3458,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3319404721260071,
            "loss": 3.932911530137062,
            "approx_kl": 0.004195764715404948,
            "wall_seconds": 2747.805160799995
          },
          {
            "update": 456,
            "environment_steps": 1867776,
            "curriculum": 3,
            "completed_episodes": 3460,
            "new_episodes": 2,
            "new_successes": 2,
            "mean_step_reward": -0.2891067862510681,
            "loss": 5.00272611528635,
            "approx_kl": 0.011539093698957004,
            "wall_seconds": 2750.5317282000033
          },
          {
            "update": 457,
            "environment_steps": 1871872,
            "curriculum": 3,
            "completed_episodes": 3462,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.31781885027885437,
            "loss": 1.9114544540643692,
            "approx_kl": 0.006585111266758759,
            "wall_seconds": 2753.2833793000027
          },
          {
            "update": 458,
            "environment_steps": 1875968,
            "curriculum": 3,
            "completed_episodes": 3469,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.31107038259506226,
            "loss": 10.857065558433533,
            "approx_kl": 0.004331242009357084,
            "wall_seconds": 2756.0022088999976
          },
          {
            "update": 459,
            "environment_steps": 1880064,
            "curriculum": 3,
            "completed_episodes": 3471,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.33399498462677,
            "loss": 2.466140441596508,
            "approx_kl": 0.009731682512210682,
            "wall_seconds": 2758.699930999981
          },
          {
            "update": 460,
            "environment_steps": 1884160,
            "curriculum": 3,
            "completed_episodes": 3477,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.29325342178344727,
            "loss": 8.793129086494446,
            "approx_kl": 0.022179131578013767,
            "wall_seconds": 2761.3895963999967
          },
          {
            "update": 461,
            "environment_steps": 1888256,
            "curriculum": 3,
            "completed_episodes": 3483,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.3242718577384949,
            "loss": 7.156977474689484,
            "approx_kl": 0.00648220350194606,
            "wall_seconds": 2764.115791999997
          },
          {
            "update": 462,
            "environment_steps": 1892352,
            "curriculum": 3,
            "completed_episodes": 3485,
            "new_episodes": 2,
            "new_successes": 2,
            "mean_step_reward": -0.2664385437965393,
            "loss": 1.9844524413347244,
            "approx_kl": 0.005617166378215188,
            "wall_seconds": 2766.8310231999785
          },
          {
            "update": 463,
            "environment_steps": 1896448,
            "curriculum": 3,
            "completed_episodes": 3493,
            "new_episodes": 8,
            "new_successes": 2,
            "mean_step_reward": -0.28311946988105774,
            "loss": 9.633182853460312,
            "approx_kl": 0.0037657374778063968,
            "wall_seconds": 2769.564733099978
          },
          {
            "update": 464,
            "environment_steps": 1900544,
            "curriculum": 3,
            "completed_episodes": 3498,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.2584766745567322,
            "loss": 8.024497717618942,
            "approx_kl": 0.0038242832633841317,
            "wall_seconds": 2772.3037375999847
          },
          {
            "update": 465,
            "environment_steps": 1904640,
            "curriculum": 3,
            "completed_episodes": 3501,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2655803859233856,
            "loss": 5.301789000630379,
            "approx_kl": 0.005088446785521228,
            "wall_seconds": 2775.041938999988
          },
          {
            "update": 466,
            "environment_steps": 1908736,
            "curriculum": 3,
            "completed_episodes": 3503,
            "new_episodes": 2,
            "new_successes": 2,
            "mean_step_reward": -0.28553658723831177,
            "loss": 2.0169631764292717,
            "approx_kl": 0.0025024912174558267,
            "wall_seconds": 2777.7576853999926
          },
          {
            "update": 467,
            "environment_steps": 1912832,
            "curriculum": 3,
            "completed_episodes": 3510,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.29620009660720825,
            "loss": 8.407513350248337,
            "approx_kl": 0.005425998573628021,
            "wall_seconds": 2780.4841462999757
          },
          {
            "update": 468,
            "environment_steps": 1916928,
            "curriculum": 3,
            "completed_episodes": 3512,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.23734119534492493,
            "loss": 4.218726322054863,
            "approx_kl": 0.004101909878954757,
            "wall_seconds": 2783.2287017999915
          },
          {
            "update": 469,
            "environment_steps": 1921024,
            "curriculum": 3,
            "completed_episodes": 3518,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.28331589698791504,
            "loss": 5.041160389780998,
            "approx_kl": 0.003095338583079865,
            "wall_seconds": 2785.9469236999867
          },
          {
            "update": 470,
            "environment_steps": 1925120,
            "curriculum": 3,
            "completed_episodes": 3522,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2907218933105469,
            "loss": 4.590126305818558,
            "approx_kl": 0.004783319280249998,
            "wall_seconds": 2788.6648708999855
          },
          {
            "update": 471,
            "environment_steps": 1929216,
            "curriculum": 3,
            "completed_episodes": 3525,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.29670587182044983,
            "loss": 2.301327880471945,
            "approx_kl": 0.003800417882303009,
            "wall_seconds": 2791.4023607999843
          },
          {
            "update": 472,
            "environment_steps": 1933312,
            "curriculum": 3,
            "completed_episodes": 3532,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2865474224090576,
            "loss": 7.2128556072711945,
            "approx_kl": 0.004554973918857286,
            "wall_seconds": 2794.1280123000033
          },
          {
            "update": 473,
            "environment_steps": 1937408,
            "curriculum": 3,
            "completed_episodes": 3538,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.30422496795654297,
            "loss": 6.870713800191879,
            "approx_kl": 0.003075359305512393,
            "wall_seconds": 2796.85408379999
          },
          {
            "update": 474,
            "environment_steps": 1941504,
            "curriculum": 3,
            "completed_episodes": 3540,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30850479006767273,
            "loss": 2.7952689304947853,
            "approx_kl": 0.00460544588349876,
            "wall_seconds": 2799.5759760999936
          },
          {
            "update": 475,
            "environment_steps": 1945600,
            "curriculum": 3,
            "completed_episodes": 3542,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.29784971475601196,
            "loss": 2.202718272805214,
            "approx_kl": 0.0032978164199448656,
            "wall_seconds": 2802.367955900001
          },
          {
            "update": 476,
            "environment_steps": 1949696,
            "curriculum": 3,
            "completed_episodes": 3545,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.30881237983703613,
            "loss": 3.7991380393505096,
            "approx_kl": 0.006969159145228332,
            "wall_seconds": 2805.084095600003
          },
          {
            "update": 477,
            "environment_steps": 1953792,
            "curriculum": 3,
            "completed_episodes": 3548,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.31857234239578247,
            "loss": 3.0631165578961372,
            "approx_kl": 0.008107529542030534,
            "wall_seconds": 2807.8208490999823
          },
          {
            "update": 478,
            "environment_steps": 1957888,
            "curriculum": 3,
            "completed_episodes": 3553,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.30287501215934753,
            "loss": 6.681859880685806,
            "approx_kl": 0.007854528237658087,
            "wall_seconds": 2810.5432898
          },
          {
            "update": 479,
            "environment_steps": 1961984,
            "curriculum": 3,
            "completed_episodes": 3556,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.31580376625061035,
            "loss": 3.6435566842556,
            "approx_kl": 0.00426607983536087,
            "wall_seconds": 2813.2918151999766
          },
          {
            "update": 480,
            "environment_steps": 1966080,
            "curriculum": 3,
            "completed_episodes": 3559,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3337806463241577,
            "loss": 2.9128442853689194,
            "approx_kl": 0.006475445792602841,
            "wall_seconds": 2815.985115599993
          },
          {
            "update": 481,
            "environment_steps": 1970176,
            "curriculum": 3,
            "completed_episodes": 3566,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.3184913992881775,
            "loss": 8.848359435796738,
            "approx_kl": 0.007215085373900365,
            "wall_seconds": 2818.7382190999924
          },
          {
            "update": 482,
            "environment_steps": 1974272,
            "curriculum": 3,
            "completed_episodes": 3571,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.2678382694721222,
            "loss": 5.147976860404015,
            "approx_kl": 0.004326416459662141,
            "wall_seconds": 2821.463040599978
          },
          {
            "update": 483,
            "environment_steps": 1978368,
            "curriculum": 3,
            "completed_episodes": 3577,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.26739513874053955,
            "loss": 6.003996059298515,
            "approx_kl": 0.003943108844396193,
            "wall_seconds": 2824.193776500004
          },
          {
            "update": 484,
            "environment_steps": 1982464,
            "curriculum": 3,
            "completed_episodes": 3581,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2792351245880127,
            "loss": 3.935814142227173,
            "approx_kl": 0.003066054232476745,
            "wall_seconds": 2826.9103017000016
          },
          {
            "update": 485,
            "environment_steps": 1986560,
            "curriculum": 3,
            "completed_episodes": 3584,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.318806529045105,
            "loss": 3.878484532237053,
            "approx_kl": 0.005606782229733653,
            "wall_seconds": 2829.632329699991
          },
          {
            "update": 486,
            "environment_steps": 1990656,
            "curriculum": 3,
            "completed_episodes": 3588,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.29807713627815247,
            "loss": 5.33406488597393,
            "approx_kl": 0.003861036544549279,
            "wall_seconds": 2832.3561774999835
          },
          {
            "update": 487,
            "environment_steps": 1994752,
            "curriculum": 3,
            "completed_episodes": 3591,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.25896981358528137,
            "loss": 3.3210902735590935,
            "approx_kl": 0.002875014495657524,
            "wall_seconds": 2835.0983832999773
          },
          {
            "update": 488,
            "environment_steps": 1998848,
            "curriculum": 3,
            "completed_episodes": 3595,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.29176464676856995,
            "loss": 3.616832658648491,
            "approx_kl": 0.0053145946803851984,
            "wall_seconds": 2837.818367400003
          },
          {
            "update": 489,
            "environment_steps": 2002944,
            "curriculum": 3,
            "completed_episodes": 3599,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3065342903137207,
            "loss": 3.3272417783737183,
            "approx_kl": 0.005993344831949798,
            "wall_seconds": 2840.5519542999973
          },
          {
            "update": 490,
            "environment_steps": 2007040,
            "curriculum": 3,
            "completed_episodes": 3601,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.27632880210876465,
            "loss": 1.9846518114209175,
            "approx_kl": 0.008446879390248796,
            "wall_seconds": 2843.280213399994
          },
          {
            "update": 491,
            "environment_steps": 2011136,
            "curriculum": 3,
            "completed_episodes": 3609,
            "new_episodes": 8,
            "new_successes": 0,
            "mean_step_reward": -0.2778550386428833,
            "loss": 10.31336396932602,
            "approx_kl": 0.0033052042163035367,
            "wall_seconds": 2846.0200688999903
          },
          {
            "update": 492,
            "environment_steps": 2015232,
            "curriculum": 3,
            "completed_episodes": 3614,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.26554811000823975,
            "loss": 5.481295883655548,
            "approx_kl": 0.004095716529263882,
            "wall_seconds": 2848.75225759999
          },
          {
            "update": 493,
            "environment_steps": 2019328,
            "curriculum": 3,
            "completed_episodes": 3620,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.3100295662879944,
            "loss": 6.548014611005783,
            "approx_kl": 0.0035072285390924662,
            "wall_seconds": 2851.476002099982
          },
          {
            "update": 494,
            "environment_steps": 2023424,
            "curriculum": 3,
            "completed_episodes": 3621,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3169691860675812,
            "loss": 1.955644890666008,
            "approx_kl": 0.006551268095790874,
            "wall_seconds": 2854.2860791999847
          },
          {
            "update": 495,
            "environment_steps": 2027520,
            "curriculum": 3,
            "completed_episodes": 3624,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3186045289039612,
            "loss": 4.084217265248299,
            "approx_kl": 0.005191818738239817,
            "wall_seconds": 2857.111873099988
          },
          {
            "update": 496,
            "environment_steps": 2031616,
            "curriculum": 3,
            "completed_episodes": 3627,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2898597717285156,
            "loss": 4.162258043885231,
            "approx_kl": 0.005847753625857877,
            "wall_seconds": 2859.9815335999883
          },
          {
            "update": 497,
            "environment_steps": 2035712,
            "curriculum": 3,
            "completed_episodes": 3631,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.31255537271499634,
            "loss": 4.630154833197594,
            "approx_kl": 0.006110209666076116,
            "wall_seconds": 2862.8310343999765
          },
          {
            "update": 498,
            "environment_steps": 2039808,
            "curriculum": 3,
            "completed_episodes": 3633,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.28768447041511536,
            "loss": 3.2188499122858047,
            "approx_kl": 0.006280882003920851,
            "wall_seconds": 2865.7704703999916
          },
          {
            "update": 499,
            "environment_steps": 2043904,
            "curriculum": 3,
            "completed_episodes": 3636,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2969478964805603,
            "loss": 2.9183055758476257,
            "approx_kl": 0.0021464311903400812,
            "wall_seconds": 2868.6996687999927
          },
          {
            "update": 500,
            "environment_steps": 2048000,
            "curriculum": 3,
            "completed_episodes": 3643,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.32962682843208313,
            "loss": 8.30828183889389,
            "approx_kl": 0.007675636617932469,
            "wall_seconds": 2871.5481669
          },
          {
            "update": 501,
            "environment_steps": 2052096,
            "curriculum": 3,
            "completed_episodes": 3648,
            "new_episodes": 5,
            "new_successes": 3,
            "mean_step_reward": -0.2712780237197876,
            "loss": 4.908612817525864,
            "approx_kl": 0.003431478082347894,
            "wall_seconds": 2874.4539736000006
          },
          {
            "update": 502,
            "environment_steps": 2056192,
            "curriculum": 3,
            "completed_episodes": 3653,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.26130378246307373,
            "loss": 7.183848887681961,
            "approx_kl": 0.0031786889230716042,
            "wall_seconds": 2877.3807722999773
          },
          {
            "update": 503,
            "environment_steps": 2060288,
            "curriculum": 3,
            "completed_episodes": 3657,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.27652376890182495,
            "loss": 4.443934887647629,
            "approx_kl": 0.003558629297913285,
            "wall_seconds": 2880.1784698999836
          },
          {
            "update": 504,
            "environment_steps": 2064384,
            "curriculum": 3,
            "completed_episodes": 3660,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.256523072719574,
            "loss": 3.2027653083205223,
            "approx_kl": 0.004208723472402198,
            "wall_seconds": 2882.94326119998
          },
          {
            "update": 505,
            "environment_steps": 2068480,
            "curriculum": 3,
            "completed_episodes": 3662,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.273703396320343,
            "loss": 3.956761136651039,
            "approx_kl": 0.004686358875915175,
            "wall_seconds": 2885.7218893999816
          },
          {
            "update": 506,
            "environment_steps": 2072576,
            "curriculum": 3,
            "completed_episodes": 3666,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.29295986890792847,
            "loss": 4.614656910300255,
            "approx_kl": 0.006014964823407354,
            "wall_seconds": 2888.594814699987
          },
          {
            "update": 507,
            "environment_steps": 2076672,
            "curriculum": 3,
            "completed_episodes": 3671,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.25451844930648804,
            "loss": 8.740466624498367,
            "approx_kl": 0.010448228040331742,
            "wall_seconds": 2891.389683699992
          },
          {
            "update": 508,
            "environment_steps": 2080768,
            "curriculum": 3,
            "completed_episodes": 3675,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2730148434638977,
            "loss": 5.2250605672597885,
            "approx_kl": 0.004308425417548278,
            "wall_seconds": 2894.2182200999814
          },
          {
            "update": 509,
            "environment_steps": 2084864,
            "curriculum": 3,
            "completed_episodes": 3678,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.31061893701553345,
            "loss": 4.296246558427811,
            "approx_kl": 0.005498748079844518,
            "wall_seconds": 2896.936568299978
          },
          {
            "update": 510,
            "environment_steps": 2088960,
            "curriculum": 3,
            "completed_episodes": 3684,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.2451585829257965,
            "loss": 8.043369978666306,
            "approx_kl": 0.004906967995339073,
            "wall_seconds": 2899.688648499985
          },
          {
            "update": 511,
            "environment_steps": 2093056,
            "curriculum": 3,
            "completed_episodes": 3689,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.20875152945518494,
            "loss": 5.302058905363083,
            "approx_kl": 0.0060278760247456376,
            "wall_seconds": 2902.457837199996
          },
          {
            "update": 512,
            "environment_steps": 2097152,
            "curriculum": 3,
            "completed_episodes": 3693,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3030623197555542,
            "loss": 3.7800983041524887,
            "approx_kl": 0.0037922314149909653,
            "wall_seconds": 2905.1809807999816
          },
          {
            "update": 513,
            "environment_steps": 2101248,
            "curriculum": 3,
            "completed_episodes": 3696,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2803962230682373,
            "loss": 2.888614811003208,
            "approx_kl": 0.005388686106016394,
            "wall_seconds": 2907.9167409999936
          },
          {
            "update": 514,
            "environment_steps": 2105344,
            "curriculum": 3,
            "completed_episodes": 3700,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2864453196525574,
            "loss": 4.48568668961525,
            "approx_kl": 0.003433933565247571,
            "wall_seconds": 2910.6913742999895
          },
          {
            "update": 515,
            "environment_steps": 2109440,
            "curriculum": 3,
            "completed_episodes": 3702,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.27549099922180176,
            "loss": 2.6547963097691536,
            "approx_kl": 0.007309621742024319,
            "wall_seconds": 2913.423539499985
          },
          {
            "update": 516,
            "environment_steps": 2113536,
            "curriculum": 3,
            "completed_episodes": 3707,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.3009679615497589,
            "loss": 3.8122363686561584,
            "approx_kl": 0.013697404148842907,
            "wall_seconds": 2916.161714499991
          },
          {
            "update": 517,
            "environment_steps": 2117632,
            "curriculum": 3,
            "completed_episodes": 3711,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.30784448981285095,
            "loss": 5.353045031428337,
            "approx_kl": 0.005904575104068499,
            "wall_seconds": 2918.882012099988
          },
          {
            "update": 518,
            "environment_steps": 2121728,
            "curriculum": 3,
            "completed_episodes": 3715,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2947007417678833,
            "loss": 4.22619691491127,
            "approx_kl": 0.002811440404911991,
            "wall_seconds": 2921.6059050999756
          },
          {
            "update": 519,
            "environment_steps": 2125824,
            "curriculum": 3,
            "completed_episodes": 3722,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.3067982792854309,
            "loss": 6.429838478565216,
            "approx_kl": 0.00413977115022135,
            "wall_seconds": 2924.343560499983
          },
          {
            "update": 520,
            "environment_steps": 2129920,
            "curriculum": 3,
            "completed_episodes": 3727,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.27543628215789795,
            "loss": 5.099015846848488,
            "approx_kl": 0.00263076583360089,
            "wall_seconds": 2927.068349599984
          },
          {
            "update": 521,
            "environment_steps": 2134016,
            "curriculum": 3,
            "completed_episodes": 3728,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.25600165128707886,
            "loss": 3.170249290764332,
            "approx_kl": 0.0025258135901822243,
            "wall_seconds": 2929.8573210999893
          },
          {
            "update": 522,
            "environment_steps": 2138112,
            "curriculum": 3,
            "completed_episodes": 3732,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2857632339000702,
            "loss": 4.442778661847115,
            "approx_kl": 0.004420791832671966,
            "wall_seconds": 2932.5918385999976
          },
          {
            "update": 523,
            "environment_steps": 2142208,
            "curriculum": 3,
            "completed_episodes": 3736,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.28602367639541626,
            "loss": 5.937721431255341,
            "approx_kl": 0.005094793781609042,
            "wall_seconds": 2935.319322700001
          },
          {
            "update": 524,
            "environment_steps": 2146304,
            "curriculum": 3,
            "completed_episodes": 3737,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.2932911515235901,
            "loss": 2.478027783334255,
            "approx_kl": 0.008782487169810338,
            "wall_seconds": 2938.0349809000036
          },
          {
            "update": 525,
            "environment_steps": 2150400,
            "curriculum": 3,
            "completed_episodes": 3741,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.31466972827911377,
            "loss": 5.262804344296455,
            "approx_kl": 0.011676928002998466,
            "wall_seconds": 2940.771210499981
          },
          {
            "update": 526,
            "environment_steps": 2154496,
            "curriculum": 3,
            "completed_episodes": 3745,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3125276565551758,
            "loss": 4.940348982810974,
            "approx_kl": 0.002739111900154967,
            "wall_seconds": 2943.48461699998
          },
          {
            "update": 527,
            "environment_steps": 2158592,
            "curriculum": 3,
            "completed_episodes": 3747,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.31359440088272095,
            "loss": 2.818250723183155,
            "approx_kl": 0.0038074409276305232,
            "wall_seconds": 2946.2153651999834
          },
          {
            "update": 528,
            "environment_steps": 2162688,
            "curriculum": 3,
            "completed_episodes": 3754,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.321745902299881,
            "loss": 7.117762416601181,
            "approx_kl": 0.007133773404348176,
            "wall_seconds": 2948.926363699982
          },
          {
            "update": 529,
            "environment_steps": 2166784,
            "curriculum": 3,
            "completed_episodes": 3759,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2699333727359772,
            "loss": 5.740353107452393,
            "approx_kl": 0.003556455332727637,
            "wall_seconds": 2951.6621410999796
          },
          {
            "update": 530,
            "environment_steps": 2170880,
            "curriculum": 3,
            "completed_episodes": 3761,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30723077058792114,
            "loss": 2.695114105939865,
            "approx_kl": 0.004526619042735547,
            "wall_seconds": 2954.378885699989
          },
          {
            "update": 531,
            "environment_steps": 2174976,
            "curriculum": 3,
            "completed_episodes": 3766,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.3166693449020386,
            "loss": 5.90169295668602,
            "approx_kl": 0.009844099138717866,
            "wall_seconds": 2957.0994912000024
          },
          {
            "update": 532,
            "environment_steps": 2179072,
            "curriculum": 3,
            "completed_episodes": 3770,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3135085105895996,
            "loss": 4.189037650823593,
            "approx_kl": 0.005618465853331145,
            "wall_seconds": 2959.8241560999886
          },
          {
            "update": 533,
            "environment_steps": 2183168,
            "curriculum": 3,
            "completed_episodes": 3772,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30644989013671875,
            "loss": 2.1428318843245506,
            "approx_kl": 0.006698451878037304,
            "wall_seconds": 2962.537274399976
          },
          {
            "update": 534,
            "environment_steps": 2187264,
            "curriculum": 3,
            "completed_episodes": 3774,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30711114406585693,
            "loss": 2.5279140546917915,
            "approx_kl": 0.005760542768257437,
            "wall_seconds": 2965.2587763999763
          },
          {
            "update": 535,
            "environment_steps": 2191360,
            "curriculum": 3,
            "completed_episodes": 3781,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.27388787269592285,
            "loss": 8.414936125278473,
            "approx_kl": 0.004614557888999116,
            "wall_seconds": 2967.9866464999795
          },
          {
            "update": 536,
            "environment_steps": 2195456,
            "curriculum": 3,
            "completed_episodes": 3783,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.26201432943344116,
            "loss": 2.7115638107061386,
            "approx_kl": 0.008784389592619846,
            "wall_seconds": 2970.7170939999924
          },
          {
            "update": 537,
            "environment_steps": 2199552,
            "curriculum": 3,
            "completed_episodes": 3786,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2732750177383423,
            "loss": 4.359734535217285,
            "approx_kl": 0.010049417353002355,
            "wall_seconds": 2973.4659484000003
          },
          {
            "update": 538,
            "environment_steps": 2203648,
            "curriculum": 3,
            "completed_episodes": 3793,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.26417675614356995,
            "loss": 9.17279314994812,
            "approx_kl": 0.0063982569190557115,
            "wall_seconds": 2976.2044161999947
          },
          {
            "update": 539,
            "environment_steps": 2207744,
            "curriculum": 3,
            "completed_episodes": 3799,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.2813113331794739,
            "loss": 6.866142779588699,
            "approx_kl": 0.004986439071217319,
            "wall_seconds": 2978.939046899992
          },
          {
            "update": 540,
            "environment_steps": 2211840,
            "curriculum": 3,
            "completed_episodes": 3801,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.28503531217575073,
            "loss": 2.0433726347982883,
            "approx_kl": 0.006484174427896505,
            "wall_seconds": 2981.671261199983
          },
          {
            "update": 541,
            "environment_steps": 2215936,
            "curriculum": 3,
            "completed_episodes": 3806,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.3015667200088501,
            "loss": 4.8878962099552155,
            "approx_kl": 0.010729900641308632,
            "wall_seconds": 2984.420322299993
          },
          {
            "update": 542,
            "environment_steps": 2220032,
            "curriculum": 3,
            "completed_episodes": 3810,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3075355291366577,
            "loss": 3.851775646209717,
            "approx_kl": 0.0050261089600098785,
            "wall_seconds": 2987.147225499997
          },
          {
            "update": 543,
            "environment_steps": 2224128,
            "curriculum": 3,
            "completed_episodes": 3811,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.30998921394348145,
            "loss": 2.2248253747820854,
            "approx_kl": 0.00392905010812683,
            "wall_seconds": 2989.9307239000045
          },
          {
            "update": 544,
            "environment_steps": 2228224,
            "curriculum": 3,
            "completed_episodes": 3814,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3141953945159912,
            "loss": 3.6003601402044296,
            "approx_kl": 0.004690767098509241,
            "wall_seconds": 2992.642361299979
          },
          {
            "update": 545,
            "environment_steps": 2232320,
            "curriculum": 3,
            "completed_episodes": 3820,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.29063689708709717,
            "loss": 5.706528306007385,
            "approx_kl": 0.006603487636311911,
            "wall_seconds": 2995.3715371
          },
          {
            "update": 546,
            "environment_steps": 2236416,
            "curriculum": 3,
            "completed_episodes": 3823,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.29564541578292847,
            "loss": 4.8778181076049805,
            "approx_kl": 0.007190854437794769,
            "wall_seconds": 2998.0843632000033
          },
          {
            "update": 547,
            "environment_steps": 2240512,
            "curriculum": 3,
            "completed_episodes": 3828,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.31283658742904663,
            "loss": 5.112917736172676,
            "approx_kl": 0.0039447191629733425,
            "wall_seconds": 3000.8107679999957
          },
          {
            "update": 548,
            "environment_steps": 2244608,
            "curriculum": 3,
            "completed_episodes": 3835,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.29799503087997437,
            "loss": 8.746940523386002,
            "approx_kl": 0.007905873688287102,
            "wall_seconds": 3003.529698899976
          },
          {
            "update": 549,
            "environment_steps": 2248704,
            "curriculum": 3,
            "completed_episodes": 3836,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.2681422233581543,
            "loss": 2.0279628708958626,
            "approx_kl": 0.00480217631047708,
            "wall_seconds": 3006.2502616999845
          },
          {
            "update": 550,
            "environment_steps": 2252800,
            "curriculum": 3,
            "completed_episodes": 3843,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.3080707788467407,
            "loss": 8.758134722709656,
            "approx_kl": 0.008275739193777554,
            "wall_seconds": 3008.9563730999944
          },
          {
            "update": 551,
            "environment_steps": 2256896,
            "curriculum": 3,
            "completed_episodes": 3845,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.31814002990722656,
            "loss": 3.4148668497800827,
            "approx_kl": 0.008894581656932132,
            "wall_seconds": 3011.7293452000013
          },
          {
            "update": 552,
            "environment_steps": 2260992,
            "curriculum": 3,
            "completed_episodes": 3846,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.30420202016830444,
            "loss": 2.010280054062605,
            "approx_kl": 0.0023130167865019757,
            "wall_seconds": 3014.470429399982
          },
          {
            "update": 553,
            "environment_steps": 2265088,
            "curriculum": 3,
            "completed_episodes": 3849,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.26355716586112976,
            "loss": 5.058044299483299,
            "approx_kl": 0.0032684061843610834,
            "wall_seconds": 3017.1956086999853
          },
          {
            "update": 554,
            "environment_steps": 2269184,
            "curriculum": 3,
            "completed_episodes": 3854,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.31618088483810425,
            "loss": 7.0573824644088745,
            "approx_kl": 0.004084842814336298,
            "wall_seconds": 3019.9264382999972
          },
          {
            "update": 555,
            "environment_steps": 2273280,
            "curriculum": 3,
            "completed_episodes": 3857,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.30680668354034424,
            "loss": 3.3036142140626907,
            "approx_kl": 0.004545103056443622,
            "wall_seconds": 3022.6534027000016
          },
          {
            "update": 556,
            "environment_steps": 2277376,
            "curriculum": 3,
            "completed_episodes": 3863,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.30775734782218933,
            "loss": 6.744641959667206,
            "approx_kl": 0.00525659095364972,
            "wall_seconds": 3025.3724870999868
          },
          {
            "update": 557,
            "environment_steps": 2281472,
            "curriculum": 3,
            "completed_episodes": 3867,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3160684108734131,
            "loss": 3.5279873609542847,
            "approx_kl": 0.005478535869769985,
            "wall_seconds": 3028.1046969000017
          },
          {
            "update": 558,
            "environment_steps": 2285568,
            "curriculum": 3,
            "completed_episodes": 3868,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3119189143180847,
            "loss": 1.4673964232206345,
            "approx_kl": 0.010102419641043525,
            "wall_seconds": 3030.8190162999963
          },
          {
            "update": 559,
            "environment_steps": 2289664,
            "curriculum": 3,
            "completed_episodes": 3873,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.26601749658584595,
            "loss": 7.32904876768589,
            "approx_kl": 0.002386374722846085,
            "wall_seconds": 3033.5420762999856
          },
          {
            "update": 560,
            "environment_steps": 2293760,
            "curriculum": 3,
            "completed_episodes": 3878,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2776831090450287,
            "loss": 6.399962693452835,
            "approx_kl": 0.0057284736139990855,
            "wall_seconds": 3036.2763737999776
          },
          {
            "update": 561,
            "environment_steps": 2297856,
            "curriculum": 3,
            "completed_episodes": 3880,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3336019814014435,
            "loss": 2.156877174973488,
            "approx_kl": 0.0058774724311660975,
            "wall_seconds": 3038.997812299989
          },
          {
            "update": 562,
            "environment_steps": 2301952,
            "curriculum": 3,
            "completed_episodes": 3884,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3116147220134735,
            "loss": 3.2152596712112427,
            "approx_kl": 0.010268957787047839,
            "wall_seconds": 3041.7036525000003
          },
          {
            "update": 563,
            "environment_steps": 2306048,
            "curriculum": 3,
            "completed_episodes": 3888,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2803308963775635,
            "loss": 5.240054190158844,
            "approx_kl": 0.006428093798604095,
            "wall_seconds": 3044.4242119999835
          },
          {
            "update": 564,
            "environment_steps": 2310144,
            "curriculum": 3,
            "completed_episodes": 3893,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.26768356561660767,
            "loss": 4.537694752216339,
            "approx_kl": 0.0062171515346562956,
            "wall_seconds": 3047.160084000003
          },
          {
            "update": 565,
            "environment_steps": 2314240,
            "curriculum": 3,
            "completed_episodes": 3897,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.30300503969192505,
            "loss": 4.569288939237595,
            "approx_kl": 0.004346844805695582,
            "wall_seconds": 3049.886431299994
          },
          {
            "update": 566,
            "environment_steps": 2318336,
            "curriculum": 3,
            "completed_episodes": 3901,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2807852029800415,
            "loss": 5.113488078117371,
            "approx_kl": 0.00435208625276573,
            "wall_seconds": 3052.6274554999836
          },
          {
            "update": 567,
            "environment_steps": 2322432,
            "curriculum": 3,
            "completed_episodes": 3906,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2720336318016052,
            "loss": 6.523788690567017,
            "approx_kl": 0.004767044585605618,
            "wall_seconds": 3055.3617231999815
          },
          {
            "update": 568,
            "environment_steps": 2326528,
            "curriculum": 3,
            "completed_episodes": 3909,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2882755696773529,
            "loss": 3.1283484995365143,
            "approx_kl": 0.0075401145113573875,
            "wall_seconds": 3058.1306893999863
          },
          {
            "update": 569,
            "environment_steps": 2330624,
            "curriculum": 3,
            "completed_episodes": 3916,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.3098025918006897,
            "loss": 8.57204058766365,
            "approx_kl": 0.003316555161291035,
            "wall_seconds": 3060.9169955999823
          },
          {
            "update": 570,
            "environment_steps": 2334720,
            "curriculum": 3,
            "completed_episodes": 3919,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.28474634885787964,
            "loss": 4.611548453569412,
            "approx_kl": 0.0041746366550796665,
            "wall_seconds": 3063.6405572999793
          },
          {
            "update": 571,
            "environment_steps": 2338816,
            "curriculum": 3,
            "completed_episodes": 3920,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.30470433831214905,
            "loss": 1.856256052851677,
            "approx_kl": 0.00922505770722637,
            "wall_seconds": 3066.374599200004
          },
          {
            "update": 572,
            "environment_steps": 2342912,
            "curriculum": 3,
            "completed_episodes": 3922,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.308637410402298,
            "loss": 3.41397824883461,
            "approx_kl": 0.0058265443112759385,
            "wall_seconds": 3069.1046135999786
          },
          {
            "update": 573,
            "environment_steps": 2347008,
            "curriculum": 3,
            "completed_episodes": 3928,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.29324641823768616,
            "loss": 5.457687735557556,
            "approx_kl": 0.006270910656894557,
            "wall_seconds": 3071.8428684999817
          },
          {
            "update": 574,
            "environment_steps": 2351104,
            "curriculum": 3,
            "completed_episodes": 3932,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2846125066280365,
            "loss": 5.164507150650024,
            "approx_kl": 0.0051102684665238485,
            "wall_seconds": 3074.5806733999925
          },
          {
            "update": 575,
            "environment_steps": 2355200,
            "curriculum": 3,
            "completed_episodes": 3936,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.25664982199668884,
            "loss": 5.288585618138313,
            "approx_kl": 0.005905161313421559,
            "wall_seconds": 3077.333249599993
          },
          {
            "update": 576,
            "environment_steps": 2359296,
            "curriculum": 3,
            "completed_episodes": 3938,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2679564356803894,
            "loss": 3.8484356850385666,
            "approx_kl": 0.004180188749160152,
            "wall_seconds": 3080.068264899979
          },
          {
            "update": 577,
            "environment_steps": 2363392,
            "curriculum": 3,
            "completed_episodes": 3942,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2860959768295288,
            "loss": 4.18841589987278,
            "approx_kl": 0.004016085324110463,
            "wall_seconds": 3082.793461699999
          },
          {
            "update": 578,
            "environment_steps": 2367488,
            "curriculum": 3,
            "completed_episodes": 3948,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.294205904006958,
            "loss": 8.465449780225754,
            "approx_kl": 0.004982498001481872,
            "wall_seconds": 3085.5286286000046
          },
          {
            "update": 579,
            "environment_steps": 2371584,
            "curriculum": 3,
            "completed_episodes": 3952,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.23838543891906738,
            "loss": 7.937422603368759,
            "approx_kl": 0.0026429455283505376,
            "wall_seconds": 3088.328278300003
          },
          {
            "update": 580,
            "environment_steps": 2375680,
            "curriculum": 3,
            "completed_episodes": 3956,
            "new_episodes": 4,
            "new_successes": 3,
            "mean_step_reward": -0.254465788602829,
            "loss": 5.027795746922493,
            "approx_kl": 0.004387956458231201,
            "wall_seconds": 3091.177100400004
          },
          {
            "update": 581,
            "environment_steps": 2379776,
            "curriculum": 3,
            "completed_episodes": 3958,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2853905260562897,
            "loss": 3.319226235151291,
            "approx_kl": 0.0036740522336913273,
            "wall_seconds": 3094.145877200004
          },
          {
            "update": 582,
            "environment_steps": 2383872,
            "curriculum": 3,
            "completed_episodes": 3964,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.30006375908851624,
            "loss": 7.332102060317993,
            "approx_kl": 0.005209060866036452,
            "wall_seconds": 3097.0388500999834
          },
          {
            "update": 583,
            "environment_steps": 2387968,
            "curriculum": 3,
            "completed_episodes": 3967,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.27849146723747253,
            "loss": 5.2047425508499146,
            "approx_kl": 0.00383209494248149,
            "wall_seconds": 3100.0308071000036
          },
          {
            "update": 584,
            "environment_steps": 2392064,
            "curriculum": 3,
            "completed_episodes": 3971,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.31335920095443726,
            "loss": 4.944451496005058,
            "approx_kl": 0.004605684505804675,
            "wall_seconds": 3102.7488934999856
          },
          {
            "update": 585,
            "environment_steps": 2396160,
            "curriculum": 3,
            "completed_episodes": 3974,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2830849587917328,
            "loss": 4.060458153486252,
            "approx_kl": 0.01977447068929905,
            "wall_seconds": 3105.469376299996
          },
          {
            "update": 586,
            "environment_steps": 2400256,
            "curriculum": 3,
            "completed_episodes": 3978,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.29565727710723877,
            "loss": 4.0932934284210205,
            "approx_kl": 0.006723795959260315,
            "wall_seconds": 3108.217148900003
          },
          {
            "update": 587,
            "environment_steps": 2404352,
            "curriculum": 3,
            "completed_episodes": 3982,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3006552755832672,
            "loss": 5.35617770254612,
            "approx_kl": 0.004567706517264014,
            "wall_seconds": 3111.0091641999898
          },
          {
            "update": 588,
            "environment_steps": 2408448,
            "curriculum": 3,
            "completed_episodes": 3987,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.30562812089920044,
            "loss": 4.0988743752241135,
            "approx_kl": 0.006020699034706922,
            "wall_seconds": 3113.740077299997
          },
          {
            "update": 589,
            "environment_steps": 2412544,
            "curriculum": 3,
            "completed_episodes": 3990,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.30227887630462646,
            "loss": 2.937403902411461,
            "approx_kl": 0.00454718742184923,
            "wall_seconds": 3116.4674071999907
          },
          {
            "update": 590,
            "environment_steps": 2416640,
            "curriculum": 3,
            "completed_episodes": 3993,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29516083002090454,
            "loss": 3.8413917869329453,
            "approx_kl": 0.00305000968000968,
            "wall_seconds": 3119.1875689000008
          },
          {
            "update": 591,
            "environment_steps": 2420736,
            "curriculum": 3,
            "completed_episodes": 3996,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.28720688819885254,
            "loss": 2.337089903652668,
            "approx_kl": 0.008043278863624437,
            "wall_seconds": 3121.9279449000023
          },
          {
            "update": 592,
            "environment_steps": 2424832,
            "curriculum": 3,
            "completed_episodes": 4001,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.27344852685928345,
            "loss": 4.912655487656593,
            "approx_kl": 0.008446971336525166,
            "wall_seconds": 3124.6586175999837
          },
          {
            "update": 593,
            "environment_steps": 2428928,
            "curriculum": 3,
            "completed_episodes": 4005,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2792324721813202,
            "loss": 5.394348815083504,
            "approx_kl": 0.0038220787937461864,
            "wall_seconds": 3127.3906471999944
          },
          {
            "update": 594,
            "environment_steps": 2433024,
            "curriculum": 3,
            "completed_episodes": 4007,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.27834248542785645,
            "loss": 2.7128773406147957,
            "approx_kl": 0.0046484328267979436,
            "wall_seconds": 3130.1111447000003
          },
          {
            "update": 595,
            "environment_steps": 2437120,
            "curriculum": 3,
            "completed_episodes": 4012,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.3025932312011719,
            "loss": 6.371823936700821,
            "approx_kl": 0.006654593558778288,
            "wall_seconds": 3132.839145999984
          },
          {
            "update": 596,
            "environment_steps": 2441216,
            "curriculum": 3,
            "completed_episodes": 4015,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29485225677490234,
            "loss": 3.698078006505966,
            "approx_kl": 0.007072065214742906,
            "wall_seconds": 3135.5724284999887
          },
          {
            "update": 597,
            "environment_steps": 2445312,
            "curriculum": 3,
            "completed_episodes": 4021,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.28822046518325806,
            "loss": 6.692009091377258,
            "approx_kl": 0.004545085470454069,
            "wall_seconds": 3138.304162699991
          },
          {
            "update": 598,
            "environment_steps": 2449408,
            "curriculum": 3,
            "completed_episodes": 4025,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2789335250854492,
            "loss": 5.014832079410553,
            "approx_kl": 0.005903039058466675,
            "wall_seconds": 3141.039832499984
          },
          {
            "update": 599,
            "environment_steps": 2453504,
            "curriculum": 3,
            "completed_episodes": 4028,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2823403775691986,
            "loss": 5.170611202716827,
            "approx_kl": 0.004907399415969849,
            "wall_seconds": 3143.7614559999784
          },
          {
            "update": 600,
            "environment_steps": 2457600,
            "curriculum": 3,
            "completed_episodes": 4033,
            "new_episodes": 5,
            "new_successes": 3,
            "mean_step_reward": -0.24539563059806824,
            "loss": 4.559895932674408,
            "approx_kl": 0.00588608274483704,
            "wall_seconds": 3146.4869362000027
          }
        ],
        "completed_episodes": 4033
      },
      "outputs/online/training-rl-gpu.json": {
        "sha256": "fc2de87efef179aaf8b95c3fca0330a209f8d88e3c0a5602c7943d319531554a",
        "metadata": {
          "method": "Joint hierarchical PPO of sensory and motor interfaces from scratch",
          "depth_renderer": "NVIDIA GeForce RTX 5090/PCIe/SSE2",
          "compute_device": "NVIDIA GeForce RTX 5090",
          "seed": 130942,
          "updates": 600,
          "environment_steps": 2457600,
          "sensory_parameters": 120577,
          "motor_parameters": 5890,
          "critic_parameters": 120577,
          "exploration_parameters": 3,
          "pretrained_neural_adapters": false,
          "anatomy": "Original fixed full retained MaleCNS graph and calibrated V1 constants; no V2/V3 weights",
          "sensory_inputs": "Fresh depth image, current XYZ and heading, goal XYZ",
          "motor_inputs": "24 PFL3 activities only, scaled by constant 10",
          "joint_log_probability": "log p(sensory cue | image, pose, goal) + log p(motor residual | PFL3 after cue)",
          "brain_gradient": "Brain transition is part of the stochastic control loop; credit assigned with policy gradients, no backpropagation into fixed anatomical weights",
          "reward": "Geodesic progress minus time/blockage penalties plus success bonus and timeout penalty; geometry is reward/physics only",
          "curriculum": "Updates 1\u201325: <=12m/25s; 26\u201350: <=30m/45s; 51\u201375: <=80m/75s; later 75% frozen training goals +25% <=80m, 120s",
          "gamma": 0.995,
          "gae_lambda": 0.95,
          "ppo_clip": 0.2,
          "entropy_coefficient": 0.004,
          "epochs": 4,
          "batch": 32,
          "horizon": 128,
          "learning_rate": 0.0003
        },
        "trace": [
          {
            "update": 1,
            "environment_steps": 4096,
            "curriculum": 0,
            "completed_episodes": 50,
            "new_episodes": 50,
            "new_successes": 50,
            "mean_step_reward": 0.08321575075387955,
            "loss": 14.524254381656647,
            "approx_kl": 0.0018040317136183148,
            "wall_seconds": 3.9911715000052936
          },
          {
            "update": 2,
            "environment_steps": 8192,
            "curriculum": 0,
            "completed_episodes": 85,
            "new_episodes": 35,
            "new_successes": 32,
            "mean_step_reward": -0.044427186250686646,
            "loss": 11.710510790348053,
            "approx_kl": 0.00047410694605787285,
            "wall_seconds": 7.443633900023997
          },
          {
            "update": 3,
            "environment_steps": 12288,
            "curriculum": 0,
            "completed_episodes": 140,
            "new_episodes": 55,
            "new_successes": 45,
            "mean_step_reward": 0.05384674668312073,
            "loss": 12.903028190135956,
            "approx_kl": 0.0006642166845267639,
            "wall_seconds": 10.937997900007758
          },
          {
            "update": 4,
            "environment_steps": 16384,
            "curriculum": 0,
            "completed_episodes": 180,
            "new_episodes": 40,
            "new_successes": 35,
            "mean_step_reward": -0.023692090064287186,
            "loss": 10.1081582903862,
            "approx_kl": 0.0018380366800556658,
            "wall_seconds": 14.411124700010987
          },
          {
            "update": 5,
            "environment_steps": 20480,
            "curriculum": 0,
            "completed_episodes": 213,
            "new_episodes": 33,
            "new_successes": 27,
            "mean_step_reward": -0.09744495153427124,
            "loss": 9.922296941280365,
            "approx_kl": 0.0008791512191237416,
            "wall_seconds": 17.8607328000071
          },
          {
            "update": 6,
            "environment_steps": 24576,
            "curriculum": 0,
            "completed_episodes": 265,
            "new_episodes": 52,
            "new_successes": 41,
            "mean_step_reward": 0.004862634465098381,
            "loss": 11.233558595180511,
            "approx_kl": 0.0013526858147088205,
            "wall_seconds": 21.344304199999897
          },
          {
            "update": 7,
            "environment_steps": 28672,
            "curriculum": 0,
            "completed_episodes": 316,
            "new_episodes": 51,
            "new_successes": 46,
            "mean_step_reward": 0.05092812329530716,
            "loss": 11.970122814178467,
            "approx_kl": 0.0017544553902553162,
            "wall_seconds": 24.849177500000224
          },
          {
            "update": 8,
            "environment_steps": 32768,
            "curriculum": 0,
            "completed_episodes": 361,
            "new_episodes": 45,
            "new_successes": 39,
            "mean_step_reward": 0.0090563353151083,
            "loss": 9.659187257289886,
            "approx_kl": 0.002619054139358923,
            "wall_seconds": 28.339232600003015
          },
          {
            "update": 9,
            "environment_steps": 36864,
            "curriculum": 0,
            "completed_episodes": 406,
            "new_episodes": 45,
            "new_successes": 37,
            "mean_step_reward": -0.01444043219089508,
            "loss": 10.68171751499176,
            "approx_kl": 0.0005714383332815487,
            "wall_seconds": 31.828859000001103
          },
          {
            "update": 10,
            "environment_steps": 40960,
            "curriculum": 0,
            "completed_episodes": 443,
            "new_episodes": 37,
            "new_successes": 30,
            "mean_step_reward": -0.06656575947999954,
            "loss": 10.099467098712921,
            "approx_kl": 0.0008886622345016804,
            "wall_seconds": 35.299867500027176
          },
          {
            "update": 11,
            "environment_steps": 45056,
            "curriculum": 0,
            "completed_episodes": 495,
            "new_episodes": 52,
            "new_successes": 45,
            "mean_step_reward": 0.06145501509308815,
            "loss": 12.106296718120575,
            "approx_kl": 0.0008757320047152461,
            "wall_seconds": 38.83999870001571
          },
          {
            "update": 12,
            "environment_steps": 49152,
            "curriculum": 0,
            "completed_episodes": 543,
            "new_episodes": 48,
            "new_successes": 40,
            "mean_step_reward": 0.011150659993290901,
            "loss": 9.736130118370056,
            "approx_kl": 0.0011193270947842393,
            "wall_seconds": 42.3195164000208
          },
          {
            "update": 13,
            "environment_steps": 53248,
            "curriculum": 0,
            "completed_episodes": 600,
            "new_episodes": 57,
            "new_successes": 49,
            "mean_step_reward": 0.08954603224992752,
            "loss": 13.0226189494133,
            "approx_kl": 0.002597654969576979,
            "wall_seconds": 45.84288970002672
          },
          {
            "update": 14,
            "environment_steps": 57344,
            "curriculum": 0,
            "completed_episodes": 643,
            "new_episodes": 43,
            "new_successes": 41,
            "mean_step_reward": 0.05422279238700867,
            "loss": 8.625183671712875,
            "approx_kl": 0.0015427618127432652,
            "wall_seconds": 49.31053430002066
          },
          {
            "update": 15,
            "environment_steps": 61440,
            "curriculum": 0,
            "completed_episodes": 685,
            "new_episodes": 42,
            "new_successes": 34,
            "mean_step_reward": -0.03519122675061226,
            "loss": 8.835269957780838,
            "approx_kl": 0.001055232249200344,
            "wall_seconds": 52.80331820002175
          },
          {
            "update": 16,
            "environment_steps": 65536,
            "curriculum": 0,
            "completed_episodes": 727,
            "new_episodes": 42,
            "new_successes": 35,
            "mean_step_reward": 0.02394082024693489,
            "loss": 12.458393394947052,
            "approx_kl": 0.0011479481618152931,
            "wall_seconds": 56.28635490001761
          },
          {
            "update": 17,
            "environment_steps": 69632,
            "curriculum": 0,
            "completed_episodes": 778,
            "new_episodes": 51,
            "new_successes": 44,
            "mean_step_reward": 0.050100475549697876,
            "loss": 9.157058030366898,
            "approx_kl": 0.0003711437984748045,
            "wall_seconds": 59.7692702000204
          },
          {
            "update": 18,
            "environment_steps": 73728,
            "curriculum": 0,
            "completed_episodes": 828,
            "new_episodes": 50,
            "new_successes": 43,
            "mean_step_reward": 0.04574849456548691,
            "loss": 9.58683705329895,
            "approx_kl": 0.0005238956546236295,
            "wall_seconds": 63.298947400006
          },
          {
            "update": 19,
            "environment_steps": 77824,
            "curriculum": 0,
            "completed_episodes": 879,
            "new_episodes": 51,
            "new_successes": 49,
            "mean_step_reward": 0.10762577503919601,
            "loss": 10.504371225833893,
            "approx_kl": 0.001830117160352529,
            "wall_seconds": 66.80227000001469
          },
          {
            "update": 20,
            "environment_steps": 81920,
            "curriculum": 0,
            "completed_episodes": 922,
            "new_episodes": 43,
            "new_successes": 40,
            "mean_step_reward": 0.04471937566995621,
            "loss": 10.721398651599884,
            "approx_kl": 0.0012676081769313896,
            "wall_seconds": 70.30239460000303
          },
          {
            "update": 21,
            "environment_steps": 86016,
            "curriculum": 0,
            "completed_episodes": 965,
            "new_episodes": 43,
            "new_successes": 34,
            "mean_step_reward": -0.034383971244096756,
            "loss": 8.711866647005081,
            "approx_kl": 0.0016179420508706244,
            "wall_seconds": 73.78263880001032
          },
          {
            "update": 22,
            "environment_steps": 90112,
            "curriculum": 0,
            "completed_episodes": 1015,
            "new_episodes": 50,
            "new_successes": 40,
            "mean_step_reward": 0.011012042872607708,
            "loss": 11.415800631046295,
            "approx_kl": 0.001840136286773486,
            "wall_seconds": 77.25502670000424
          },
          {
            "update": 23,
            "environment_steps": 94208,
            "curriculum": 0,
            "completed_episodes": 1066,
            "new_episodes": 51,
            "new_successes": 44,
            "mean_step_reward": 0.037162259221076965,
            "loss": 11.982712388038635,
            "approx_kl": 0.0005913618497288553,
            "wall_seconds": 80.75145880001946
          },
          {
            "update": 24,
            "environment_steps": 98304,
            "curriculum": 0,
            "completed_episodes": 1104,
            "new_episodes": 38,
            "new_successes": 33,
            "mean_step_reward": -0.030383534729480743,
            "loss": 10.37479054927826,
            "approx_kl": 0.0015808558582648402,
            "wall_seconds": 84.22647360002156
          },
          {
            "update": 25,
            "environment_steps": 102400,
            "curriculum": 0,
            "completed_episodes": 1152,
            "new_episodes": 48,
            "new_successes": 38,
            "mean_step_reward": 0.02812524139881134,
            "loss": 12.389942586421967,
            "approx_kl": 0.0024759703628660645,
            "wall_seconds": 87.75378430000274
          },
          {
            "update": 26,
            "environment_steps": 106496,
            "curriculum": 1,
            "completed_episodes": 1181,
            "new_episodes": 29,
            "new_successes": 29,
            "mean_step_reward": 0.017578165978193283,
            "loss": 8.713993638753891,
            "approx_kl": 0.0008743080506974366,
            "wall_seconds": 91.23434650001582
          },
          {
            "update": 27,
            "environment_steps": 110592,
            "curriculum": 1,
            "completed_episodes": 1211,
            "new_episodes": 30,
            "new_successes": 27,
            "mean_step_reward": -0.01191866397857666,
            "loss": 7.163691729307175,
            "approx_kl": 0.0017758588055585278,
            "wall_seconds": 94.70580030002748
          },
          {
            "update": 28,
            "environment_steps": 114688,
            "curriculum": 1,
            "completed_episodes": 1234,
            "new_episodes": 23,
            "new_successes": 20,
            "mean_step_reward": -0.08866795152425766,
            "loss": 7.9598002433776855,
            "approx_kl": 0.0013016317516303388,
            "wall_seconds": 98.18934350000927
          },
          {
            "update": 29,
            "environment_steps": 118784,
            "curriculum": 1,
            "completed_episodes": 1252,
            "new_episodes": 18,
            "new_successes": 14,
            "mean_step_reward": -0.13271576166152954,
            "loss": 9.487732231616974,
            "approx_kl": 0.0017690464210318169,
            "wall_seconds": 101.61956720001763
          },
          {
            "update": 30,
            "environment_steps": 122880,
            "curriculum": 1,
            "completed_episodes": 1273,
            "new_episodes": 21,
            "new_successes": 14,
            "mean_step_reward": -0.17778989672660828,
            "loss": 8.310596734285355,
            "approx_kl": 0.003378743684152141,
            "wall_seconds": 105.05029470002046
          },
          {
            "update": 31,
            "environment_steps": 126976,
            "curriculum": 1,
            "completed_episodes": 1287,
            "new_episodes": 14,
            "new_successes": 10,
            "mean_step_reward": -0.1711869239807129,
            "loss": 7.467989265918732,
            "approx_kl": 0.0007326927861868171,
            "wall_seconds": 108.55137340002693
          },
          {
            "update": 32,
            "environment_steps": 131072,
            "curriculum": 1,
            "completed_episodes": 1307,
            "new_episodes": 20,
            "new_successes": 8,
            "mean_step_reward": -0.21702057123184204,
            "loss": 7.292920112609863,
            "approx_kl": 0.002877585064197774,
            "wall_seconds": 112.00244680000469
          },
          {
            "update": 33,
            "environment_steps": 135168,
            "curriculum": 1,
            "completed_episodes": 1329,
            "new_episodes": 22,
            "new_successes": 18,
            "mean_step_reward": -0.13610653579235077,
            "loss": 8.734405010938644,
            "approx_kl": 0.0009224125005857786,
            "wall_seconds": 115.45353760002763
          },
          {
            "update": 34,
            "environment_steps": 139264,
            "curriculum": 1,
            "completed_episodes": 1349,
            "new_episodes": 20,
            "new_successes": 14,
            "mean_step_reward": -0.1365736722946167,
            "loss": 9.62924239039421,
            "approx_kl": 0.002765733934211312,
            "wall_seconds": 118.90528220002307
          },
          {
            "update": 35,
            "environment_steps": 143360,
            "curriculum": 1,
            "completed_episodes": 1359,
            "new_episodes": 10,
            "new_successes": 6,
            "mean_step_reward": -0.24738167226314545,
            "loss": 5.7034448981285095,
            "approx_kl": 0.0010853468938876176,
            "wall_seconds": 122.36485640000319
          },
          {
            "update": 36,
            "environment_steps": 147456,
            "curriculum": 1,
            "completed_episodes": 1375,
            "new_episodes": 16,
            "new_successes": 8,
            "mean_step_reward": -0.2058843970298767,
            "loss": 8.933372348546982,
            "approx_kl": 0.0029060228443995584,
            "wall_seconds": 125.77931350001018
          },
          {
            "update": 37,
            "environment_steps": 151552,
            "curriculum": 1,
            "completed_episodes": 1393,
            "new_episodes": 18,
            "new_successes": 10,
            "mean_step_reward": -0.20044267177581787,
            "loss": 8.555563926696777,
            "approx_kl": 0.0006847731438028859,
            "wall_seconds": 129.21755500001018
          },
          {
            "update": 38,
            "environment_steps": 155648,
            "curriculum": 1,
            "completed_episodes": 1426,
            "new_episodes": 33,
            "new_successes": 23,
            "mean_step_reward": -0.056406617164611816,
            "loss": 18.513587832450867,
            "approx_kl": 0.0030013308769412106,
            "wall_seconds": 132.71028930001194
          },
          {
            "update": 39,
            "environment_steps": 159744,
            "curriculum": 1,
            "completed_episodes": 1448,
            "new_episodes": 22,
            "new_successes": 19,
            "mean_step_reward": -0.09399737417697906,
            "loss": 15.84001350402832,
            "approx_kl": 0.001258619407963124,
            "wall_seconds": 136.15948530001333
          },
          {
            "update": 40,
            "environment_steps": 163840,
            "curriculum": 1,
            "completed_episodes": 1465,
            "new_episodes": 17,
            "new_successes": 10,
            "mean_step_reward": -0.18140441179275513,
            "loss": 10.291739702224731,
            "approx_kl": 0.0015027210774860578,
            "wall_seconds": 139.58640290002222
          },
          {
            "update": 41,
            "environment_steps": 167936,
            "curriculum": 1,
            "completed_episodes": 1493,
            "new_episodes": 28,
            "new_successes": 21,
            "mean_step_reward": -0.0876576378941536,
            "loss": 12.843280136585236,
            "approx_kl": 0.001632439874811098,
            "wall_seconds": 143.086518600001
          },
          {
            "update": 42,
            "environment_steps": 172032,
            "curriculum": 1,
            "completed_episodes": 1512,
            "new_episodes": 19,
            "new_successes": 14,
            "mean_step_reward": -0.16118812561035156,
            "loss": 10.55837219953537,
            "approx_kl": 0.0014421192845475161,
            "wall_seconds": 146.5346357000235
          },
          {
            "update": 43,
            "environment_steps": 176128,
            "curriculum": 1,
            "completed_episodes": 1535,
            "new_episodes": 23,
            "new_successes": 18,
            "mean_step_reward": -0.10814939439296722,
            "loss": 10.574401199817657,
            "approx_kl": 0.0029716664139414206,
            "wall_seconds": 149.9749279000098
          },
          {
            "update": 44,
            "environment_steps": 180224,
            "curriculum": 1,
            "completed_episodes": 1554,
            "new_episodes": 19,
            "new_successes": 14,
            "mean_step_reward": -0.17159003019332886,
            "loss": 9.103024244308472,
            "approx_kl": 0.0011923381662199972,
            "wall_seconds": 153.46373500002665
          },
          {
            "update": 45,
            "environment_steps": 184320,
            "curriculum": 1,
            "completed_episodes": 1568,
            "new_episodes": 14,
            "new_successes": 7,
            "mean_step_reward": -0.23765316605567932,
            "loss": 6.795748949050903,
            "approx_kl": 0.002267279342049733,
            "wall_seconds": 156.91456970002037
          },
          {
            "update": 46,
            "environment_steps": 188416,
            "curriculum": 1,
            "completed_episodes": 1586,
            "new_episodes": 18,
            "new_successes": 12,
            "mean_step_reward": -0.1787053942680359,
            "loss": 12.877140164375305,
            "approx_kl": 0.0018157857175538084,
            "wall_seconds": 160.3516921000264
          },
          {
            "update": 47,
            "environment_steps": 192512,
            "curriculum": 1,
            "completed_episodes": 1602,
            "new_episodes": 16,
            "new_successes": 11,
            "mean_step_reward": -0.17389574646949768,
            "loss": 9.259571850299835,
            "approx_kl": 0.0014520486274705036,
            "wall_seconds": 163.81826480000746
          },
          {
            "update": 48,
            "environment_steps": 196608,
            "curriculum": 1,
            "completed_episodes": 1628,
            "new_episodes": 26,
            "new_successes": 16,
            "mean_step_reward": -0.1253577470779419,
            "loss": 12.790173530578613,
            "approx_kl": 0.0015889575934124878,
            "wall_seconds": 167.28951640002197
          },
          {
            "update": 49,
            "environment_steps": 200704,
            "curriculum": 1,
            "completed_episodes": 1652,
            "new_episodes": 24,
            "new_successes": 20,
            "mean_step_reward": -0.0673709437251091,
            "loss": 14.587607502937317,
            "approx_kl": 0.0018123195077350829,
            "wall_seconds": 170.75952080002753
          },
          {
            "update": 50,
            "environment_steps": 204800,
            "curriculum": 1,
            "completed_episodes": 1681,
            "new_episodes": 29,
            "new_successes": 25,
            "mean_step_reward": -0.05740093067288399,
            "loss": 13.372774720191956,
            "approx_kl": 0.00397364439231751,
            "wall_seconds": 174.23538550001103
          },
          {
            "update": 51,
            "environment_steps": 208896,
            "curriculum": 2,
            "completed_episodes": 1693,
            "new_episodes": 12,
            "new_successes": 12,
            "mean_step_reward": -0.1543155014514923,
            "loss": 8.495191097259521,
            "approx_kl": 0.003599308287448366,
            "wall_seconds": 177.72686429999885
          },
          {
            "update": 52,
            "environment_steps": 212992,
            "curriculum": 2,
            "completed_episodes": 1699,
            "new_episodes": 6,
            "new_successes": 6,
            "mean_step_reward": -0.22064107656478882,
            "loss": 8.350239843130112,
            "approx_kl": 0.0016959260174189694,
            "wall_seconds": 181.13813210002263
          },
          {
            "update": 53,
            "environment_steps": 217088,
            "curriculum": 2,
            "completed_episodes": 1703,
            "new_episodes": 4,
            "new_successes": 4,
            "mean_step_reward": -0.2705959677696228,
            "loss": 4.777160748839378,
            "approx_kl": 0.001858245024777716,
            "wall_seconds": 184.55554580001626
          },
          {
            "update": 54,
            "environment_steps": 221184,
            "curriculum": 2,
            "completed_episodes": 1713,
            "new_episodes": 10,
            "new_successes": 3,
            "mean_step_reward": -0.2964155673980713,
            "loss": 5.081621050834656,
            "approx_kl": 0.002411305018540588,
            "wall_seconds": 188.00588720000815
          },
          {
            "update": 55,
            "environment_steps": 225280,
            "curriculum": 2,
            "completed_episodes": 1718,
            "new_episodes": 5,
            "new_successes": 3,
            "mean_step_reward": -0.2654453217983246,
            "loss": 3.0211226791143417,
            "approx_kl": 0.0018131941615138203,
            "wall_seconds": 191.39524520002306
          },
          {
            "update": 56,
            "environment_steps": 229376,
            "curriculum": 2,
            "completed_episodes": 1727,
            "new_episodes": 9,
            "new_successes": 2,
            "mean_step_reward": -0.25718045234680176,
            "loss": 4.932700932025909,
            "approx_kl": 0.0015950719825923443,
            "wall_seconds": 194.83271750001586
          },
          {
            "update": 57,
            "environment_steps": 233472,
            "curriculum": 2,
            "completed_episodes": 1736,
            "new_episodes": 9,
            "new_successes": 4,
            "mean_step_reward": -0.2531253695487976,
            "loss": 4.861414760351181,
            "approx_kl": 0.0029837616621080087,
            "wall_seconds": 198.25485740002478
          },
          {
            "update": 58,
            "environment_steps": 237568,
            "curriculum": 2,
            "completed_episodes": 1747,
            "new_episodes": 11,
            "new_successes": 7,
            "mean_step_reward": -0.2321508526802063,
            "loss": 6.302436411380768,
            "approx_kl": 0.0030646657633042196,
            "wall_seconds": 201.72180590001517
          },
          {
            "update": 59,
            "environment_steps": 241664,
            "curriculum": 2,
            "completed_episodes": 1754,
            "new_episodes": 7,
            "new_successes": 3,
            "mean_step_reward": -0.2771880030632019,
            "loss": 5.7335003316402435,
            "approx_kl": 0.0032037614782893797,
            "wall_seconds": 205.1496432000131
          },
          {
            "update": 60,
            "environment_steps": 245760,
            "curriculum": 2,
            "completed_episodes": 1766,
            "new_episodes": 12,
            "new_successes": 6,
            "mean_step_reward": -0.227461576461792,
            "loss": 8.736228555440903,
            "approx_kl": 0.003616007172240643,
            "wall_seconds": 208.58941280000727
          },
          {
            "update": 61,
            "environment_steps": 249856,
            "curriculum": 2,
            "completed_episodes": 1771,
            "new_episodes": 5,
            "new_successes": 3,
            "mean_step_reward": -0.27213841676712036,
            "loss": 4.597179725766182,
            "approx_kl": 0.004247368895448744,
            "wall_seconds": 212.04460140000447
          },
          {
            "update": 62,
            "environment_steps": 253952,
            "curriculum": 2,
            "completed_episodes": 1779,
            "new_episodes": 8,
            "new_successes": 1,
            "mean_step_reward": -0.29073184728622437,
            "loss": 5.68312007188797,
            "approx_kl": 0.002746614251009305,
            "wall_seconds": 215.46910220000427
          },
          {
            "update": 63,
            "environment_steps": 258048,
            "curriculum": 2,
            "completed_episodes": 1789,
            "new_episodes": 10,
            "new_successes": 6,
            "mean_step_reward": -0.21619051694869995,
            "loss": 9.628166228532791,
            "approx_kl": 0.0026651175176084507,
            "wall_seconds": 218.9199185000034
          },
          {
            "update": 64,
            "environment_steps": 262144,
            "curriculum": 2,
            "completed_episodes": 1796,
            "new_episodes": 7,
            "new_successes": 4,
            "mean_step_reward": -0.2621998190879822,
            "loss": 5.320423275232315,
            "approx_kl": 0.003825387357210275,
            "wall_seconds": 222.3559073000215
          },
          {
            "update": 65,
            "environment_steps": 266240,
            "curriculum": 2,
            "completed_episodes": 1800,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.32830357551574707,
            "loss": 2.141734555363655,
            "approx_kl": 0.003938861082133371,
            "wall_seconds": 225.7396137000178
          },
          {
            "update": 66,
            "environment_steps": 270336,
            "curriculum": 2,
            "completed_episodes": 1812,
            "new_episodes": 12,
            "new_successes": 2,
            "mean_step_reward": -0.30281901359558105,
            "loss": 5.421329826116562,
            "approx_kl": 0.004905260500891018,
            "wall_seconds": 229.16676500000176
          },
          {
            "update": 67,
            "environment_steps": 274432,
            "curriculum": 2,
            "completed_episodes": 1817,
            "new_episodes": 5,
            "new_successes": 3,
            "mean_step_reward": -0.24902385473251343,
            "loss": 5.599637031555176,
            "approx_kl": 0.007702175584199722,
            "wall_seconds": 232.58030330002657
          },
          {
            "update": 68,
            "environment_steps": 278528,
            "curriculum": 2,
            "completed_episodes": 1824,
            "new_episodes": 7,
            "new_successes": 4,
            "mean_step_reward": -0.2656809687614441,
            "loss": 7.768030971288681,
            "approx_kl": 0.0034559926425572485,
            "wall_seconds": 235.99656290002167
          },
          {
            "update": 69,
            "environment_steps": 282624,
            "curriculum": 2,
            "completed_episodes": 1832,
            "new_episodes": 8,
            "new_successes": 1,
            "mean_step_reward": -0.2773974537849426,
            "loss": 8.292327284812927,
            "approx_kl": 0.0014304023643489927,
            "wall_seconds": 239.4408717000042
          },
          {
            "update": 70,
            "environment_steps": 286720,
            "curriculum": 2,
            "completed_episodes": 1839,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.2912006974220276,
            "loss": 5.4949155151844025,
            "approx_kl": 0.002791218013953767,
            "wall_seconds": 242.85634260001825
          },
          {
            "update": 71,
            "environment_steps": 290816,
            "curriculum": 2,
            "completed_episodes": 1843,
            "new_episodes": 4,
            "new_successes": 3,
            "mean_step_reward": -0.27115434408187866,
            "loss": 5.680626422166824,
            "approx_kl": 0.005047829208706389,
            "wall_seconds": 246.3137687999988
          },
          {
            "update": 72,
            "environment_steps": 294912,
            "curriculum": 2,
            "completed_episodes": 1852,
            "new_episodes": 9,
            "new_successes": 2,
            "mean_step_reward": -0.24855858087539673,
            "loss": 8.13269305229187,
            "approx_kl": 0.006026387271049316,
            "wall_seconds": 249.77948710002238
          },
          {
            "update": 73,
            "environment_steps": 299008,
            "curriculum": 2,
            "completed_episodes": 1859,
            "new_episodes": 7,
            "new_successes": 3,
            "mean_step_reward": -0.2579202950000763,
            "loss": 8.967909157276154,
            "approx_kl": 0.0023581468594784383,
            "wall_seconds": 253.19517030002316
          },
          {
            "update": 74,
            "environment_steps": 303104,
            "curriculum": 2,
            "completed_episodes": 1865,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.2169235795736313,
            "loss": 7.063971430063248,
            "approx_kl": 0.0039058330385159934,
            "wall_seconds": 256.6234779000224
          },
          {
            "update": 75,
            "environment_steps": 307200,
            "curriculum": 2,
            "completed_episodes": 1883,
            "new_episodes": 18,
            "new_successes": 12,
            "mean_step_reward": -0.10322452336549759,
            "loss": 17.593556821346283,
            "approx_kl": 0.002255723744383431,
            "wall_seconds": 260.1176322000101
          },
          {
            "update": 76,
            "environment_steps": 311296,
            "curriculum": 3,
            "completed_episodes": 1886,
            "new_episodes": 3,
            "new_successes": 3,
            "mean_step_reward": -0.2105274796485901,
            "loss": 8.158747911453247,
            "approx_kl": 0.001737796777888434,
            "wall_seconds": 263.5567389000207
          },
          {
            "update": 77,
            "environment_steps": 315392,
            "curriculum": 3,
            "completed_episodes": 1888,
            "new_episodes": 2,
            "new_successes": 2,
            "mean_step_reward": -0.274962842464447,
            "loss": 4.018760234117508,
            "approx_kl": 0.003097951353993267,
            "wall_seconds": 266.9607067000179
          },
          {
            "update": 78,
            "environment_steps": 319488,
            "curriculum": 3,
            "completed_episodes": 1888,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.32878512144088745,
            "loss": 0.8378270789980888,
            "approx_kl": 0.005373560521547915,
            "wall_seconds": 270.3234885999991
          },
          {
            "update": 79,
            "environment_steps": 323584,
            "curriculum": 3,
            "completed_episodes": 1889,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3281288743019104,
            "loss": 1.3791238442063332,
            "approx_kl": 0.003947567643990624,
            "wall_seconds": 273.6840734000143
          },
          {
            "update": 80,
            "environment_steps": 327680,
            "curriculum": 3,
            "completed_episodes": 1891,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.3107730746269226,
            "loss": 4.5756309032440186,
            "approx_kl": 0.017273898161874968,
            "wall_seconds": 277.0751085000229
          },
          {
            "update": 81,
            "environment_steps": 331776,
            "curriculum": 3,
            "completed_episodes": 1895,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2987934947013855,
            "loss": 4.878043532371521,
            "approx_kl": 0.0356779534558882,
            "wall_seconds": 280.49517380000907
          },
          {
            "update": 82,
            "environment_steps": 335872,
            "curriculum": 3,
            "completed_episodes": 1899,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2901080250740051,
            "loss": 3.635482504963875,
            "approx_kl": 0.011763671061999048,
            "wall_seconds": 283.9190865000128
          },
          {
            "update": 83,
            "environment_steps": 339968,
            "curriculum": 3,
            "completed_episodes": 1906,
            "new_episodes": 7,
            "new_successes": 3,
            "mean_step_reward": -0.28776687383651733,
            "loss": 7.8667764365673065,
            "approx_kl": 0.004820338290301152,
            "wall_seconds": 287.3238087000209
          },
          {
            "update": 84,
            "environment_steps": 344064,
            "curriculum": 3,
            "completed_episodes": 1915,
            "new_episodes": 9,
            "new_successes": 1,
            "mean_step_reward": -0.27350395917892456,
            "loss": 8.594762116670609,
            "approx_kl": 0.0064257137764798244,
            "wall_seconds": 290.7548648000229
          },
          {
            "update": 85,
            "environment_steps": 348160,
            "curriculum": 3,
            "completed_episodes": 1926,
            "new_episodes": 11,
            "new_successes": 3,
            "mean_step_reward": -0.27165457606315613,
            "loss": 12.162007510662079,
            "approx_kl": 0.0031439085378224263,
            "wall_seconds": 294.21293910001987
          },
          {
            "update": 86,
            "environment_steps": 352256,
            "curriculum": 3,
            "completed_episodes": 1926,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.3007858395576477,
            "loss": 1.5285764560103416,
            "approx_kl": 0.0027040253444283735,
            "wall_seconds": 297.61895830000867
          },
          {
            "update": 87,
            "environment_steps": 356352,
            "curriculum": 3,
            "completed_episodes": 1927,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.32152727246284485,
            "loss": 1.13802694901824,
            "approx_kl": 0.009934470575899468,
            "wall_seconds": 300.9963437000115
          },
          {
            "update": 88,
            "environment_steps": 360448,
            "curriculum": 3,
            "completed_episodes": 1927,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.30907541513442993,
            "loss": 1.2319513745605946,
            "approx_kl": 0.00439552373609331,
            "wall_seconds": 304.37119760000496
          },
          {
            "update": 89,
            "environment_steps": 364544,
            "curriculum": 3,
            "completed_episodes": 1929,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3279002606868744,
            "loss": 1.7485083937644958,
            "approx_kl": 0.016138975208377815,
            "wall_seconds": 307.74100110001746
          },
          {
            "update": 90,
            "environment_steps": 368640,
            "curriculum": 3,
            "completed_episodes": 1931,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.324901282787323,
            "loss": 2.265931434929371,
            "approx_kl": 0.004177787344815442,
            "wall_seconds": 311.12380820000544
          },
          {
            "update": 91,
            "environment_steps": 372736,
            "curriculum": 3,
            "completed_episodes": 1935,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.33152616024017334,
            "loss": 3.7136542797088623,
            "approx_kl": 0.0050842783548432635,
            "wall_seconds": 314.5858977000171
          },
          {
            "update": 92,
            "environment_steps": 376832,
            "curriculum": 3,
            "completed_episodes": 1940,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.30735793709754944,
            "loss": 5.8525799214839935,
            "approx_kl": 0.0057665010517666815,
            "wall_seconds": 317.9893250000023
          },
          {
            "update": 93,
            "environment_steps": 380928,
            "curriculum": 3,
            "completed_episodes": 1946,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.2883913516998291,
            "loss": 9.200474947690964,
            "approx_kl": 0.003865205091642565,
            "wall_seconds": 321.3971975000168
          },
          {
            "update": 94,
            "environment_steps": 385024,
            "curriculum": 3,
            "completed_episodes": 1956,
            "new_episodes": 10,
            "new_successes": 2,
            "mean_step_reward": -0.28181976079940796,
            "loss": 10.836353480815887,
            "approx_kl": 0.002695644450795953,
            "wall_seconds": 324.8358366000175
          },
          {
            "update": 95,
            "environment_steps": 389120,
            "curriculum": 3,
            "completed_episodes": 1963,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.283441960811615,
            "loss": 7.161699831485748,
            "approx_kl": 0.002673381084605353,
            "wall_seconds": 328.296259499999
          },
          {
            "update": 96,
            "environment_steps": 393216,
            "curriculum": 3,
            "completed_episodes": 1965,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2968568801879883,
            "loss": 2.23539862036705,
            "approx_kl": 0.005081425419120933,
            "wall_seconds": 331.6994651000132
          },
          {
            "update": 97,
            "environment_steps": 397312,
            "curriculum": 3,
            "completed_episodes": 1967,
            "new_episodes": 2,
            "new_successes": 2,
            "mean_step_reward": -0.2725702226161957,
            "loss": 2.516410931944847,
            "approx_kl": 0.0037772127416246803,
            "wall_seconds": 335.10826550002093
          },
          {
            "update": 98,
            "environment_steps": 401408,
            "curriculum": 3,
            "completed_episodes": 1970,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2805716097354889,
            "loss": 6.089309722185135,
            "approx_kl": 0.00515035661192087,
            "wall_seconds": 338.5352328000008
          },
          {
            "update": 99,
            "environment_steps": 405504,
            "curriculum": 3,
            "completed_episodes": 1972,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.31970325112342834,
            "loss": 1.8924988359212875,
            "approx_kl": 0.0038419137490564026,
            "wall_seconds": 341.9304293000023
          },
          {
            "update": 100,
            "environment_steps": 409600,
            "curriculum": 3,
            "completed_episodes": 1976,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.31527233123779297,
            "loss": 4.992401614785194,
            "approx_kl": 0.007253862813740852,
            "wall_seconds": 345.32417750000604
          },
          {
            "update": 101,
            "environment_steps": 413696,
            "curriculum": 3,
            "completed_episodes": 1979,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2964063286781311,
            "loss": 2.324492707848549,
            "approx_kl": 0.0026864696646953234,
            "wall_seconds": 348.7816070000117
          },
          {
            "update": 102,
            "environment_steps": 417792,
            "curriculum": 3,
            "completed_episodes": 1985,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.3044137954711914,
            "loss": 7.966989487409592,
            "approx_kl": 0.004271806104952702,
            "wall_seconds": 352.23064870000235
          },
          {
            "update": 103,
            "environment_steps": 421888,
            "curriculum": 3,
            "completed_episodes": 1996,
            "new_episodes": 11,
            "new_successes": 4,
            "mean_step_reward": -0.2598540186882019,
            "loss": 10.876268208026886,
            "approx_kl": 0.0034830704043997684,
            "wall_seconds": 355.68839570001
          },
          {
            "update": 104,
            "environment_steps": 425984,
            "curriculum": 3,
            "completed_episodes": 2005,
            "new_episodes": 9,
            "new_successes": 4,
            "mean_step_reward": -0.20315077900886536,
            "loss": 10.045109212398529,
            "approx_kl": 0.002336490682864678,
            "wall_seconds": 359.1123214000254
          },
          {
            "update": 105,
            "environment_steps": 430080,
            "curriculum": 3,
            "completed_episodes": 2009,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2526257634162903,
            "loss": 6.6274756491184235,
            "approx_kl": 0.0030453877989202738,
            "wall_seconds": 362.52822560002096
          },
          {
            "update": 106,
            "environment_steps": 434176,
            "curriculum": 3,
            "completed_episodes": 2013,
            "new_episodes": 4,
            "new_successes": 4,
            "mean_step_reward": -0.23586563766002655,
            "loss": 4.149889439344406,
            "approx_kl": 0.004939189941069344,
            "wall_seconds": 365.9755698000081
          },
          {
            "update": 107,
            "environment_steps": 438272,
            "curriculum": 3,
            "completed_episodes": 2016,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2950839400291443,
            "loss": 4.716212972998619,
            "approx_kl": 0.004793753751073382,
            "wall_seconds": 369.4088049000129
          },
          {
            "update": 108,
            "environment_steps": 442368,
            "curriculum": 3,
            "completed_episodes": 2017,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3141513764858246,
            "loss": 2.3426002636551857,
            "approx_kl": 0.00554683395603206,
            "wall_seconds": 373.44604460001574
          },
          {
            "update": 109,
            "environment_steps": 446464,
            "curriculum": 3,
            "completed_episodes": 2021,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.26804351806640625,
            "loss": 6.819557815790176,
            "approx_kl": 0.0017188148649438517,
            "wall_seconds": 379.9919306000229
          },
          {
            "update": 110,
            "environment_steps": 450560,
            "curriculum": 3,
            "completed_episodes": 2025,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.26444011926651,
            "loss": 7.2720354199409485,
            "approx_kl": 0.0026908658328466117,
            "wall_seconds": 386.6424837000086
          },
          {
            "update": 111,
            "environment_steps": 454656,
            "curriculum": 3,
            "completed_episodes": 2029,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.24191617965698242,
            "loss": 4.693931132555008,
            "approx_kl": 0.0029046071249467786,
            "wall_seconds": 393.37950020001153
          },
          {
            "update": 112,
            "environment_steps": 458752,
            "curriculum": 3,
            "completed_episodes": 2034,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2822205424308777,
            "loss": 7.678934633731842,
            "approx_kl": 0.0036353272171254503,
            "wall_seconds": 400.16002030001255
          },
          {
            "update": 113,
            "environment_steps": 462848,
            "curriculum": 3,
            "completed_episodes": 2042,
            "new_episodes": 8,
            "new_successes": 1,
            "mean_step_reward": -0.2855774760246277,
            "loss": 11.030960381031036,
            "approx_kl": 0.0034533574216766283,
            "wall_seconds": 406.8312256000063
          },
          {
            "update": 114,
            "environment_steps": 466944,
            "curriculum": 3,
            "completed_episodes": 2046,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.273538202047348,
            "loss": 5.555978432297707,
            "approx_kl": 0.003982543816164252,
            "wall_seconds": 413.558619700023
          },
          {
            "update": 115,
            "environment_steps": 471040,
            "curriculum": 3,
            "completed_episodes": 2050,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2811620533466339,
            "loss": 3.798982709646225,
            "approx_kl": 0.004433791797055164,
            "wall_seconds": 420.2162820000085
          },
          {
            "update": 116,
            "environment_steps": 475136,
            "curriculum": 3,
            "completed_episodes": 2052,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.31186550855636597,
            "loss": 3.3282690048217773,
            "approx_kl": 0.004517562198088854,
            "wall_seconds": 425.2805820000067
          },
          {
            "update": 117,
            "environment_steps": 479232,
            "curriculum": 3,
            "completed_episodes": 2054,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.3048378825187683,
            "loss": 3.6861321330070496,
            "approx_kl": 0.006445198985602474,
            "wall_seconds": 428.6846863000246
          },
          {
            "update": 118,
            "environment_steps": 483328,
            "curriculum": 3,
            "completed_episodes": 2056,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3124128580093384,
            "loss": 2.2466549575328827,
            "approx_kl": 0.010357123959693126,
            "wall_seconds": 432.09106890001567
          },
          {
            "update": 119,
            "environment_steps": 487424,
            "curriculum": 3,
            "completed_episodes": 2061,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2824147343635559,
            "loss": 8.287589222192764,
            "approx_kl": 0.00570398095260316,
            "wall_seconds": 435.5560424000141
          },
          {
            "update": 120,
            "environment_steps": 491520,
            "curriculum": 3,
            "completed_episodes": 2065,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.29735374450683594,
            "loss": 5.658622473478317,
            "approx_kl": 0.006531506880492088,
            "wall_seconds": 438.972004500014
          },
          {
            "update": 121,
            "environment_steps": 495616,
            "curriculum": 3,
            "completed_episodes": 2068,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2916469871997833,
            "loss": 3.687557503581047,
            "approx_kl": 0.008013422222575173,
            "wall_seconds": 442.41743330002646
          },
          {
            "update": 122,
            "environment_steps": 499712,
            "curriculum": 3,
            "completed_episodes": 2076,
            "new_episodes": 8,
            "new_successes": 0,
            "mean_step_reward": -0.2871549129486084,
            "loss": 10.072706133127213,
            "approx_kl": 0.006428036373108625,
            "wall_seconds": 445.87958860001527
          },
          {
            "update": 123,
            "environment_steps": 503808,
            "curriculum": 3,
            "completed_episodes": 2082,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.21486540138721466,
            "loss": 9.164151728153229,
            "approx_kl": 0.002891440784878796,
            "wall_seconds": 449.3402629000193
          },
          {
            "update": 124,
            "environment_steps": 507904,
            "curriculum": 3,
            "completed_episodes": 2084,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2647024393081665,
            "loss": 3.4020635336637497,
            "approx_kl": 0.0031756247881276067,
            "wall_seconds": 452.78184940002393
          },
          {
            "update": 125,
            "environment_steps": 512000,
            "curriculum": 3,
            "completed_episodes": 2088,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2912754416465759,
            "loss": 9.42508602142334,
            "approx_kl": 0.0022673748062516097,
            "wall_seconds": 456.408101900015
          },
          {
            "update": 126,
            "environment_steps": 516096,
            "curriculum": 3,
            "completed_episodes": 2090,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.30776873230934143,
            "loss": 2.9762897193431854,
            "approx_kl": 0.0030470228666672483,
            "wall_seconds": 459.8325112000166
          },
          {
            "update": 127,
            "environment_steps": 520192,
            "curriculum": 3,
            "completed_episodes": 2091,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.28310099244117737,
            "loss": 2.708220236003399,
            "approx_kl": 0.005810329637824907,
            "wall_seconds": 463.310358000017
          },
          {
            "update": 128,
            "environment_steps": 524288,
            "curriculum": 3,
            "completed_episodes": 2095,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.29736626148223877,
            "loss": 4.0461007952690125,
            "approx_kl": 0.003915203611541074,
            "wall_seconds": 466.72948420001194
          },
          {
            "update": 129,
            "environment_steps": 528384,
            "curriculum": 3,
            "completed_episodes": 2100,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2692118287086487,
            "loss": 5.14303332567215,
            "approx_kl": 0.004028781884699129,
            "wall_seconds": 470.17069280002033
          },
          {
            "update": 130,
            "environment_steps": 532480,
            "curriculum": 3,
            "completed_episodes": 2103,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.21729762852191925,
            "loss": 7.048615038394928,
            "approx_kl": 0.00533652599551715,
            "wall_seconds": 473.63896229999955
          },
          {
            "update": 131,
            "environment_steps": 536576,
            "curriculum": 3,
            "completed_episodes": 2110,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.26640164852142334,
            "loss": 7.361255645751953,
            "approx_kl": 0.006228502912563272,
            "wall_seconds": 477.11616180001874
          },
          {
            "update": 132,
            "environment_steps": 540672,
            "curriculum": 3,
            "completed_episodes": 2119,
            "new_episodes": 9,
            "new_successes": 2,
            "mean_step_reward": -0.2227674126625061,
            "loss": 9.250304639339447,
            "approx_kl": 0.005392036995544913,
            "wall_seconds": 480.5640418000112
          },
          {
            "update": 133,
            "environment_steps": 544768,
            "curriculum": 3,
            "completed_episodes": 2122,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.24359658360481262,
            "loss": 3.7838612347841263,
            "approx_kl": 0.003699435303133214,
            "wall_seconds": 483.9889457000245
          },
          {
            "update": 134,
            "environment_steps": 548864,
            "curriculum": 3,
            "completed_episodes": 2127,
            "new_episodes": 5,
            "new_successes": 3,
            "mean_step_reward": -0.24538332223892212,
            "loss": 9.373366296291351,
            "approx_kl": 0.0034444731900293846,
            "wall_seconds": 487.4701474000176
          },
          {
            "update": 135,
            "environment_steps": 552960,
            "curriculum": 3,
            "completed_episodes": 2130,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2680943012237549,
            "loss": 6.148366272449493,
            "approx_kl": 0.004400557243570802,
            "wall_seconds": 490.8935043000092
          },
          {
            "update": 136,
            "environment_steps": 557056,
            "curriculum": 3,
            "completed_episodes": 2133,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.27524083852767944,
            "loss": 4.388418227434158,
            "approx_kl": 0.00501971220728592,
            "wall_seconds": 494.3121068000037
          },
          {
            "update": 137,
            "environment_steps": 561152,
            "curriculum": 3,
            "completed_episodes": 2135,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.311332106590271,
            "loss": 3.3768108785152435,
            "approx_kl": 0.004517299279541476,
            "wall_seconds": 497.7433324000158
          },
          {
            "update": 138,
            "environment_steps": 565248,
            "curriculum": 3,
            "completed_episodes": 2138,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3147972524166107,
            "loss": 3.2081083953380585,
            "approx_kl": 0.0040678989917068975,
            "wall_seconds": 501.1542875000159
          },
          {
            "update": 139,
            "environment_steps": 569344,
            "curriculum": 3,
            "completed_episodes": 2142,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3128637671470642,
            "loss": 5.642514139413834,
            "approx_kl": 0.003448256526098703,
            "wall_seconds": 504.562287800014
          },
          {
            "update": 140,
            "environment_steps": 573440,
            "curriculum": 3,
            "completed_episodes": 2147,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.3113182783126831,
            "loss": 5.5619703233242035,
            "approx_kl": 0.005346127101802267,
            "wall_seconds": 507.9798800999997
          },
          {
            "update": 141,
            "environment_steps": 577536,
            "curriculum": 3,
            "completed_episodes": 2154,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.2977495789527893,
            "loss": 9.439000278711319,
            "approx_kl": 0.003380416319487267,
            "wall_seconds": 511.460779200017
          },
          {
            "update": 142,
            "environment_steps": 581632,
            "curriculum": 3,
            "completed_episodes": 2156,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3098602592945099,
            "loss": 2.6972388699650764,
            "approx_kl": 0.004812129991478287,
            "wall_seconds": 514.8797517000057
          },
          {
            "update": 143,
            "environment_steps": 585728,
            "curriculum": 3,
            "completed_episodes": 2163,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.28223973512649536,
            "loss": 9.514928162097931,
            "approx_kl": 0.004801670093002031,
            "wall_seconds": 518.3233094000025
          },
          {
            "update": 144,
            "environment_steps": 589824,
            "curriculum": 3,
            "completed_episodes": 2166,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.27277427911758423,
            "loss": 4.3475561290979385,
            "approx_kl": 0.002626303028591792,
            "wall_seconds": 521.7774593000067
          },
          {
            "update": 145,
            "environment_steps": 593920,
            "curriculum": 3,
            "completed_episodes": 2169,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.277580589056015,
            "loss": 3.6597305089235306,
            "approx_kl": 0.004972564405761659,
            "wall_seconds": 525.4392712000117
          },
          {
            "update": 146,
            "environment_steps": 598016,
            "curriculum": 3,
            "completed_episodes": 2172,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.31080421805381775,
            "loss": 4.754165977239609,
            "approx_kl": 0.004683722807385493,
            "wall_seconds": 528.8819615000102
          },
          {
            "update": 147,
            "environment_steps": 602112,
            "curriculum": 3,
            "completed_episodes": 2175,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.28536635637283325,
            "loss": 4.3709767162799835,
            "approx_kl": 0.004725334554677829,
            "wall_seconds": 532.3072413000045
          },
          {
            "update": 148,
            "environment_steps": 606208,
            "curriculum": 3,
            "completed_episodes": 2179,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3112597167491913,
            "loss": 4.092593729496002,
            "approx_kl": 0.007521082619859953,
            "wall_seconds": 535.7363682000141
          },
          {
            "update": 149,
            "environment_steps": 610304,
            "curriculum": 3,
            "completed_episodes": 2183,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.32861703634262085,
            "loss": 4.525895223021507,
            "approx_kl": 0.007305049331989721,
            "wall_seconds": 539.1579853000003
          },
          {
            "update": 150,
            "environment_steps": 614400,
            "curriculum": 3,
            "completed_episodes": 2187,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3025805950164795,
            "loss": 5.1628019362688065,
            "approx_kl": 0.0056218932768388186,
            "wall_seconds": 542.5727406000078
          },
          {
            "update": 151,
            "environment_steps": 618496,
            "curriculum": 3,
            "completed_episodes": 2191,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.29326730966567993,
            "loss": 5.073289483785629,
            "approx_kl": 0.004605419186191284,
            "wall_seconds": 545.9982438000152
          },
          {
            "update": 152,
            "environment_steps": 622592,
            "curriculum": 3,
            "completed_episodes": 2197,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.2983948588371277,
            "loss": 6.511977463960648,
            "approx_kl": 0.003198455267920508,
            "wall_seconds": 549.4675101000175
          },
          {
            "update": 153,
            "environment_steps": 626688,
            "curriculum": 3,
            "completed_episodes": 2202,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.2537458539009094,
            "loss": 8.008105903863907,
            "approx_kl": 0.003937218218197813,
            "wall_seconds": 552.914534200012
          },
          {
            "update": 154,
            "environment_steps": 630784,
            "curriculum": 3,
            "completed_episodes": 2207,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.2541786730289459,
            "loss": 5.56757926940918,
            "approx_kl": 0.002905109819039353,
            "wall_seconds": 556.3286041000101
          },
          {
            "update": 155,
            "environment_steps": 634880,
            "curriculum": 3,
            "completed_episodes": 2209,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2732018828392029,
            "loss": 6.891331851482391,
            "approx_kl": 0.005025050484618987,
            "wall_seconds": 559.7556972000166
          },
          {
            "update": 156,
            "environment_steps": 638976,
            "curriculum": 3,
            "completed_episodes": 2213,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.27258169651031494,
            "loss": 6.060238242149353,
            "approx_kl": 0.004514593398198485,
            "wall_seconds": 563.1982377000095
          },
          {
            "update": 157,
            "environment_steps": 643072,
            "curriculum": 3,
            "completed_episodes": 2214,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.27284225821495056,
            "loss": 2.5959616899490356,
            "approx_kl": 0.007561250053186086,
            "wall_seconds": 566.6623512000078
          },
          {
            "update": 158,
            "environment_steps": 647168,
            "curriculum": 3,
            "completed_episodes": 2218,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.326576828956604,
            "loss": 5.763882517814636,
            "approx_kl": 0.006002633246680489,
            "wall_seconds": 570.0699338000268
          },
          {
            "update": 159,
            "environment_steps": 651264,
            "curriculum": 3,
            "completed_episodes": 2224,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.30282682180404663,
            "loss": 7.944898992776871,
            "approx_kl": 0.0026885161805694224,
            "wall_seconds": 573.4744390000124
          },
          {
            "update": 160,
            "environment_steps": 655360,
            "curriculum": 3,
            "completed_episodes": 2229,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.28906452655792236,
            "loss": 4.743468195199966,
            "approx_kl": 0.003413143454963574,
            "wall_seconds": 576.9089373000024
          },
          {
            "update": 161,
            "environment_steps": 659456,
            "curriculum": 3,
            "completed_episodes": 2233,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.27725714445114136,
            "loss": 4.818156450986862,
            "approx_kl": 0.004296591199818067,
            "wall_seconds": 580.3645189000235
          },
          {
            "update": 162,
            "environment_steps": 663552,
            "curriculum": 3,
            "completed_episodes": 2238,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.265855073928833,
            "loss": 7.309650868177414,
            "approx_kl": 0.0044246315919735935,
            "wall_seconds": 583.8040126000124
          },
          {
            "update": 163,
            "environment_steps": 667648,
            "curriculum": 3,
            "completed_episodes": 2242,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2744535207748413,
            "loss": 5.194407224655151,
            "approx_kl": 0.00313056746199436,
            "wall_seconds": 587.2290119000245
          },
          {
            "update": 164,
            "environment_steps": 671744,
            "curriculum": 3,
            "completed_episodes": 2247,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.2998332977294922,
            "loss": 2.127191260457039,
            "approx_kl": 0.006243947398616001,
            "wall_seconds": 590.6564553000208
          },
          {
            "update": 165,
            "environment_steps": 675840,
            "curriculum": 3,
            "completed_episodes": 2251,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.31655263900756836,
            "loss": 2.689551681280136,
            "approx_kl": 0.010076204296638025,
            "wall_seconds": 594.0595223000273
          },
          {
            "update": 166,
            "environment_steps": 679936,
            "curriculum": 3,
            "completed_episodes": 2254,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3051387667655945,
            "loss": 3.351624608039856,
            "approx_kl": 0.006113549083238468,
            "wall_seconds": 597.4708733000152
          },
          {
            "update": 167,
            "environment_steps": 684032,
            "curriculum": 3,
            "completed_episodes": 2256,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3143892288208008,
            "loss": 3.010172486305237,
            "approx_kl": 0.008535744545952184,
            "wall_seconds": 600.8725515000115
          },
          {
            "update": 168,
            "environment_steps": 688128,
            "curriculum": 3,
            "completed_episodes": 2259,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2878149747848511,
            "loss": 3.9554616063833237,
            "approx_kl": 0.015280865904060192,
            "wall_seconds": 604.3012954000151
          },
          {
            "update": 169,
            "environment_steps": 692224,
            "curriculum": 3,
            "completed_episodes": 2266,
            "new_episodes": 7,
            "new_successes": 3,
            "mean_step_reward": -0.25087451934814453,
            "loss": 10.007272392511368,
            "approx_kl": 0.009284258496336406,
            "wall_seconds": 607.7628473000077
          },
          {
            "update": 170,
            "environment_steps": 696320,
            "curriculum": 3,
            "completed_episodes": 2271,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.26719507575035095,
            "loss": 7.956018358469009,
            "approx_kl": 0.004990671843188466,
            "wall_seconds": 611.1909048000234
          },
          {
            "update": 171,
            "environment_steps": 700416,
            "curriculum": 3,
            "completed_episodes": 2275,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2854481041431427,
            "loss": 6.427225053310394,
            "approx_kl": 0.0023738333857181715,
            "wall_seconds": 614.6495772000053
          },
          {
            "update": 172,
            "environment_steps": 704512,
            "curriculum": 3,
            "completed_episodes": 2281,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.26557621359825134,
            "loss": 6.52293273806572,
            "approx_kl": 0.00232060420421476,
            "wall_seconds": 618.1103485000203
          },
          {
            "update": 173,
            "environment_steps": 708608,
            "curriculum": 3,
            "completed_episodes": 2285,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3041455149650574,
            "loss": 4.965317457914352,
            "approx_kl": 0.003432353021707968,
            "wall_seconds": 621.5174212999991
          },
          {
            "update": 174,
            "environment_steps": 712704,
            "curriculum": 3,
            "completed_episodes": 2287,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.31970852613449097,
            "loss": 2.4644878432154655,
            "approx_kl": 0.003954284327846835,
            "wall_seconds": 625.080407100002
          },
          {
            "update": 175,
            "environment_steps": 716800,
            "curriculum": 3,
            "completed_episodes": 2291,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.30836495757102966,
            "loss": 4.9125242084264755,
            "approx_kl": 0.0021687967964680865,
            "wall_seconds": 628.5236197000195
          },
          {
            "update": 176,
            "environment_steps": 720896,
            "curriculum": 3,
            "completed_episodes": 2292,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.2902919054031372,
            "loss": 1.7887777462601662,
            "approx_kl": 0.00581075859554403,
            "wall_seconds": 631.9521219000162
          },
          {
            "update": 177,
            "environment_steps": 724992,
            "curriculum": 3,
            "completed_episodes": 2296,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2670896649360657,
            "loss": 5.291965439915657,
            "approx_kl": 0.005020001297452836,
            "wall_seconds": 635.4011893999996
          },
          {
            "update": 178,
            "environment_steps": 729088,
            "curriculum": 3,
            "completed_episodes": 2303,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2873837947845459,
            "loss": 6.765825271606445,
            "approx_kl": 0.0047229576211975655,
            "wall_seconds": 638.838276800001
          },
          {
            "update": 179,
            "environment_steps": 733184,
            "curriculum": 3,
            "completed_episodes": 2305,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2658096253871918,
            "loss": 4.226031377911568,
            "approx_kl": 0.00308176951148198,
            "wall_seconds": 642.2563539000112
          },
          {
            "update": 180,
            "environment_steps": 737280,
            "curriculum": 3,
            "completed_episodes": 2311,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.30951225757598877,
            "loss": 6.511741071939468,
            "approx_kl": 0.0035877125810657162,
            "wall_seconds": 645.6796195000061
          },
          {
            "update": 181,
            "environment_steps": 741376,
            "curriculum": 3,
            "completed_episodes": 2316,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.22684311866760254,
            "loss": 6.370030373334885,
            "approx_kl": 0.0035271678898425307,
            "wall_seconds": 649.2115930000145
          },
          {
            "update": 182,
            "environment_steps": 745472,
            "curriculum": 3,
            "completed_episodes": 2319,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.21051615476608276,
            "loss": 5.7273156344890594,
            "approx_kl": 0.0022518470923387213,
            "wall_seconds": 652.6937426000077
          },
          {
            "update": 183,
            "environment_steps": 749568,
            "curriculum": 3,
            "completed_episodes": 2321,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.23976103961467743,
            "loss": 3.420894607901573,
            "approx_kl": 0.0037425135742523707,
            "wall_seconds": 656.1197590000229
          },
          {
            "update": 184,
            "environment_steps": 753664,
            "curriculum": 3,
            "completed_episodes": 2324,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29567426443099976,
            "loss": 4.259313613176346,
            "approx_kl": 0.003182959571859101,
            "wall_seconds": 659.5719335000031
          },
          {
            "update": 185,
            "environment_steps": 757760,
            "curriculum": 3,
            "completed_episodes": 2324,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.263211727142334,
            "loss": 1.6356296092271805,
            "approx_kl": 0.004840665895244456,
            "wall_seconds": 662.9957713000185
          },
          {
            "update": 186,
            "environment_steps": 761856,
            "curriculum": 3,
            "completed_episodes": 2328,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.29958486557006836,
            "loss": 4.71056792140007,
            "approx_kl": 0.0048462079612363596,
            "wall_seconds": 666.4105444000161
          },
          {
            "update": 187,
            "environment_steps": 765952,
            "curriculum": 3,
            "completed_episodes": 2332,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2977858781814575,
            "loss": 5.241477727890015,
            "approx_kl": 0.004509808562943363,
            "wall_seconds": 669.9053780000249
          },
          {
            "update": 188,
            "environment_steps": 770048,
            "curriculum": 3,
            "completed_episodes": 2337,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.25970539450645447,
            "loss": 4.01622362434864,
            "approx_kl": 0.005327405686330167,
            "wall_seconds": 673.458528600022
          },
          {
            "update": 189,
            "environment_steps": 774144,
            "curriculum": 3,
            "completed_episodes": 2342,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2911929786205292,
            "loss": 5.613080620765686,
            "approx_kl": 0.0052244272319512675,
            "wall_seconds": 676.9377781000221
          },
          {
            "update": 190,
            "environment_steps": 778240,
            "curriculum": 3,
            "completed_episodes": 2346,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.33133000135421753,
            "loss": 3.939587488770485,
            "approx_kl": 0.0033763396731956163,
            "wall_seconds": 680.3431819000107
          },
          {
            "update": 191,
            "environment_steps": 782336,
            "curriculum": 3,
            "completed_episodes": 2353,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2995684742927551,
            "loss": 6.824246019124985,
            "approx_kl": 0.005189968915146892,
            "wall_seconds": 683.7842908000166
          },
          {
            "update": 192,
            "environment_steps": 786432,
            "curriculum": 3,
            "completed_episodes": 2357,
            "new_episodes": 4,
            "new_successes": 3,
            "mean_step_reward": -0.21369494497776031,
            "loss": 7.764601230621338,
            "approx_kl": 0.005310321223078063,
            "wall_seconds": 687.2603163000022
          },
          {
            "update": 193,
            "environment_steps": 790528,
            "curriculum": 3,
            "completed_episodes": 2360,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2532736361026764,
            "loss": 4.289738401770592,
            "approx_kl": 0.004540024439847912,
            "wall_seconds": 690.7308027000108
          },
          {
            "update": 194,
            "environment_steps": 794624,
            "curriculum": 3,
            "completed_episodes": 2365,
            "new_episodes": 5,
            "new_successes": 4,
            "mean_step_reward": -0.21430185437202454,
            "loss": 9.076527148485184,
            "approx_kl": 0.007158200724006747,
            "wall_seconds": 694.1977526000119
          },
          {
            "update": 195,
            "environment_steps": 798720,
            "curriculum": 3,
            "completed_episodes": 2367,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2906690835952759,
            "loss": 2.3846049085259438,
            "approx_kl": 0.004106886312001734,
            "wall_seconds": 697.6251893000153
          },
          {
            "update": 196,
            "environment_steps": 802816,
            "curriculum": 3,
            "completed_episodes": 2372,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.3168209493160248,
            "loss": 5.358093947172165,
            "approx_kl": 0.0028020551308145514,
            "wall_seconds": 701.0421466000262
          },
          {
            "update": 197,
            "environment_steps": 806912,
            "curriculum": 3,
            "completed_episodes": 2377,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2960421144962311,
            "loss": 7.338777005672455,
            "approx_kl": 0.004565820163406897,
            "wall_seconds": 704.4482747000002
          },
          {
            "update": 198,
            "environment_steps": 811008,
            "curriculum": 3,
            "completed_episodes": 2382,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2893174886703491,
            "loss": 5.493945851922035,
            "approx_kl": 0.0031829533363634255,
            "wall_seconds": 707.9040838000074
          },
          {
            "update": 199,
            "environment_steps": 815104,
            "curriculum": 3,
            "completed_episodes": 2385,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.29521334171295166,
            "loss": 3.087974525988102,
            "approx_kl": 0.0033844033005152596,
            "wall_seconds": 711.310603800026
          },
          {
            "update": 200,
            "environment_steps": 819200,
            "curriculum": 3,
            "completed_episodes": 2388,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.31741005182266235,
            "loss": 3.719943881034851,
            "approx_kl": 0.0024690737682249164,
            "wall_seconds": 714.7029977000202
          },
          {
            "update": 201,
            "environment_steps": 823296,
            "curriculum": 3,
            "completed_episodes": 2392,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2668308615684509,
            "loss": 5.207925111055374,
            "approx_kl": 0.001777631841832772,
            "wall_seconds": 718.1582435000164
          },
          {
            "update": 202,
            "environment_steps": 827392,
            "curriculum": 3,
            "completed_episodes": 2394,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2907051742076874,
            "loss": 3.0492726266384125,
            "approx_kl": 0.0031730000191600993,
            "wall_seconds": 721.5979559000116
          },
          {
            "update": 203,
            "environment_steps": 831488,
            "curriculum": 3,
            "completed_episodes": 2398,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.313071608543396,
            "loss": 4.220329135656357,
            "approx_kl": 0.0034466260294720996,
            "wall_seconds": 724.9957087000075
          },
          {
            "update": 204,
            "environment_steps": 835584,
            "curriculum": 3,
            "completed_episodes": 2401,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2819408178329468,
            "loss": 5.0126578360795975,
            "approx_kl": 0.005239455413175165,
            "wall_seconds": 728.4289735000057
          },
          {
            "update": 205,
            "environment_steps": 839680,
            "curriculum": 3,
            "completed_episodes": 2405,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.26364201307296753,
            "loss": 4.31019289791584,
            "approx_kl": 0.0049079723885370186,
            "wall_seconds": 731.9271744000143
          },
          {
            "update": 206,
            "environment_steps": 843776,
            "curriculum": 3,
            "completed_episodes": 2412,
            "new_episodes": 7,
            "new_successes": 3,
            "mean_step_reward": -0.2406567633152008,
            "loss": 8.338905185461044,
            "approx_kl": 0.007360913694355986,
            "wall_seconds": 735.3674662000267
          },
          {
            "update": 207,
            "environment_steps": 847872,
            "curriculum": 3,
            "completed_episodes": 2418,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.23719929158687592,
            "loss": 5.57111468911171,
            "approx_kl": 0.0032679247560736258,
            "wall_seconds": 738.8327482000168
          },
          {
            "update": 208,
            "environment_steps": 851968,
            "curriculum": 3,
            "completed_episodes": 2422,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2843111753463745,
            "loss": 5.084724098443985,
            "approx_kl": 0.003930851918994449,
            "wall_seconds": 742.260531500011
          },
          {
            "update": 209,
            "environment_steps": 856064,
            "curriculum": 3,
            "completed_episodes": 2424,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2811499834060669,
            "loss": 3.2772802859544754,
            "approx_kl": 0.0037214105450402712,
            "wall_seconds": 745.7243559000199
          },
          {
            "update": 210,
            "environment_steps": 860160,
            "curriculum": 3,
            "completed_episodes": 2428,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2985894978046417,
            "loss": 4.402633115649223,
            "approx_kl": 0.004636586412743782,
            "wall_seconds": 749.1411551000201
          },
          {
            "update": 211,
            "environment_steps": 864256,
            "curriculum": 3,
            "completed_episodes": 2432,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2529248893260956,
            "loss": 7.7899512350559235,
            "approx_kl": 0.004523953646639711,
            "wall_seconds": 752.5623729000217
          },
          {
            "update": 212,
            "environment_steps": 868352,
            "curriculum": 3,
            "completed_episodes": 2437,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.28311729431152344,
            "loss": 5.534558549523354,
            "approx_kl": 0.0032688261326256907,
            "wall_seconds": 755.9643810000271
          },
          {
            "update": 213,
            "environment_steps": 872448,
            "curriculum": 3,
            "completed_episodes": 2439,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.29950958490371704,
            "loss": 3.87643101811409,
            "approx_kl": 0.003527946591930231,
            "wall_seconds": 759.3896506000019
          },
          {
            "update": 214,
            "environment_steps": 876544,
            "curriculum": 3,
            "completed_episodes": 2443,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.29801738262176514,
            "loss": 4.236529111862183,
            "approx_kl": 0.00408597975547309,
            "wall_seconds": 762.8129631000047
          },
          {
            "update": 215,
            "environment_steps": 880640,
            "curriculum": 3,
            "completed_episodes": 2447,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.30735108256340027,
            "loss": 5.251518443226814,
            "approx_kl": 0.0021488411621248815,
            "wall_seconds": 766.2449680000136
          },
          {
            "update": 216,
            "environment_steps": 884736,
            "curriculum": 3,
            "completed_episodes": 2452,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2743813991546631,
            "loss": 5.386496990919113,
            "approx_kl": 0.005274521667161025,
            "wall_seconds": 769.6930139000178
          },
          {
            "update": 217,
            "environment_steps": 888832,
            "curriculum": 3,
            "completed_episodes": 2458,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.27826958894729614,
            "loss": 6.469686686992645,
            "approx_kl": 0.004164782896623365,
            "wall_seconds": 773.1039375000109
          },
          {
            "update": 218,
            "environment_steps": 892928,
            "curriculum": 3,
            "completed_episodes": 2462,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2908537983894348,
            "loss": 4.128405898809433,
            "approx_kl": 0.0023196202746476047,
            "wall_seconds": 776.5720693000185
          },
          {
            "update": 219,
            "environment_steps": 897024,
            "curriculum": 3,
            "completed_episodes": 2466,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.30209457874298096,
            "loss": 7.475580006837845,
            "approx_kl": 0.004830494621273829,
            "wall_seconds": 779.9997756000084
          },
          {
            "update": 220,
            "environment_steps": 901120,
            "curriculum": 3,
            "completed_episodes": 2470,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2802610397338867,
            "loss": 5.7665261924266815,
            "approx_kl": 0.004146354600379709,
            "wall_seconds": 783.4217994000064
          },
          {
            "update": 221,
            "environment_steps": 905216,
            "curriculum": 3,
            "completed_episodes": 2472,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.29578980803489685,
            "loss": 3.7493798434734344,
            "approx_kl": 0.004153164467425086,
            "wall_seconds": 786.907321000006
          },
          {
            "update": 222,
            "environment_steps": 909312,
            "curriculum": 3,
            "completed_episodes": 2474,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2793371081352234,
            "loss": 2.9545941203832626,
            "approx_kl": 0.0036854181525995955,
            "wall_seconds": 790.3596227000235
          },
          {
            "update": 223,
            "environment_steps": 913408,
            "curriculum": 3,
            "completed_episodes": 2478,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3022124767303467,
            "loss": 4.785075068473816,
            "approx_kl": 0.003451413900620537,
            "wall_seconds": 793.7720089000068
          },
          {
            "update": 224,
            "environment_steps": 917504,
            "curriculum": 3,
            "completed_episodes": 2482,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.33351337909698486,
            "loss": 3.81877239048481,
            "approx_kl": 0.006435451950892457,
            "wall_seconds": 797.1730211000249
          },
          {
            "update": 225,
            "environment_steps": 921600,
            "curriculum": 3,
            "completed_episodes": 2487,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.29926052689552307,
            "loss": 6.067000508308411,
            "approx_kl": 0.005313764111633645,
            "wall_seconds": 800.6121000000276
          },
          {
            "update": 226,
            "environment_steps": 925696,
            "curriculum": 3,
            "completed_episodes": 2492,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.3032933473587036,
            "loss": 6.500259190797806,
            "approx_kl": 0.004315418251280789,
            "wall_seconds": 804.0350384000049
          },
          {
            "update": 227,
            "environment_steps": 929792,
            "curriculum": 3,
            "completed_episodes": 2495,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3233816623687744,
            "loss": 3.476160705089569,
            "approx_kl": 0.007018298980256077,
            "wall_seconds": 807.4304470000206
          },
          {
            "update": 228,
            "environment_steps": 933888,
            "curriculum": 3,
            "completed_episodes": 2497,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3329193592071533,
            "loss": 1.9477684274315834,
            "approx_kl": 0.005357656062187743,
            "wall_seconds": 810.8433187000046
          },
          {
            "update": 229,
            "environment_steps": 937984,
            "curriculum": 3,
            "completed_episodes": 2501,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3162155747413635,
            "loss": 3.396850898861885,
            "approx_kl": 0.004822737579161185,
            "wall_seconds": 814.2715709000186
          },
          {
            "update": 230,
            "environment_steps": 942080,
            "curriculum": 3,
            "completed_episodes": 2504,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3175904452800751,
            "loss": 3.250114291906357,
            "approx_kl": 0.007839544407033827,
            "wall_seconds": 817.7160484000051
          },
          {
            "update": 231,
            "environment_steps": 946176,
            "curriculum": 3,
            "completed_episodes": 2507,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2733457386493683,
            "loss": 4.934397026896477,
            "approx_kl": 0.007910009677289054,
            "wall_seconds": 821.1582153000054
          },
          {
            "update": 232,
            "environment_steps": 950272,
            "curriculum": 3,
            "completed_episodes": 2512,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.3007131516933441,
            "loss": 7.432646065950394,
            "approx_kl": 0.005258919874904677,
            "wall_seconds": 824.5852767000033
          },
          {
            "update": 233,
            "environment_steps": 954368,
            "curriculum": 3,
            "completed_episodes": 2514,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.29565536975860596,
            "loss": 3.5099355429410934,
            "approx_kl": 0.0049299490510748,
            "wall_seconds": 828.0387498000055
          },
          {
            "update": 234,
            "environment_steps": 958464,
            "curriculum": 3,
            "completed_episodes": 2522,
            "new_episodes": 8,
            "new_successes": 2,
            "mean_step_reward": -0.2521563470363617,
            "loss": 10.952821373939514,
            "approx_kl": 0.00544780998461647,
            "wall_seconds": 831.4542450000008
          },
          {
            "update": 235,
            "environment_steps": 962560,
            "curriculum": 3,
            "completed_episodes": 2527,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.22655437886714935,
            "loss": 7.483419328927994,
            "approx_kl": 0.003202080477421987,
            "wall_seconds": 834.906979899999
          },
          {
            "update": 236,
            "environment_steps": 966656,
            "curriculum": 3,
            "completed_episodes": 2533,
            "new_episodes": 6,
            "new_successes": 3,
            "mean_step_reward": -0.23289597034454346,
            "loss": 8.453874260187149,
            "approx_kl": 0.0030207479085220257,
            "wall_seconds": 838.3843340000021
          },
          {
            "update": 237,
            "environment_steps": 970752,
            "curriculum": 3,
            "completed_episodes": 2536,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2932951748371124,
            "loss": 3.8540591448545456,
            "approx_kl": 0.0037552322828560136,
            "wall_seconds": 841.8086747000052
          },
          {
            "update": 238,
            "environment_steps": 974848,
            "curriculum": 3,
            "completed_episodes": 2540,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.32206663489341736,
            "loss": 3.383710503578186,
            "approx_kl": 0.0038842261837999104,
            "wall_seconds": 845.2163892000099
          },
          {
            "update": 239,
            "environment_steps": 978944,
            "curriculum": 3,
            "completed_episodes": 2541,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.307517945766449,
            "loss": 2.2596092373132706,
            "approx_kl": 0.0040449281405017246,
            "wall_seconds": 848.6240773000172
          },
          {
            "update": 240,
            "environment_steps": 983040,
            "curriculum": 3,
            "completed_episodes": 2544,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3176509141921997,
            "loss": 3.7349823564291,
            "approx_kl": 0.004604020326951286,
            "wall_seconds": 852.0361537999997
          },
          {
            "update": 241,
            "environment_steps": 987136,
            "curriculum": 3,
            "completed_episodes": 2548,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2791242003440857,
            "loss": 5.3826853185892105,
            "approx_kl": 0.003079386930039618,
            "wall_seconds": 855.5218892000266
          },
          {
            "update": 242,
            "environment_steps": 991232,
            "curriculum": 3,
            "completed_episodes": 2553,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.24648824334144592,
            "loss": 4.571912914514542,
            "approx_kl": 0.004992436777683906,
            "wall_seconds": 858.9658504000108
          },
          {
            "update": 243,
            "environment_steps": 995328,
            "curriculum": 3,
            "completed_episodes": 2557,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2986605167388916,
            "loss": 4.85500006377697,
            "approx_kl": 0.003848579293844523,
            "wall_seconds": 862.3868249000225
          },
          {
            "update": 244,
            "environment_steps": 999424,
            "curriculum": 3,
            "completed_episodes": 2560,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29535675048828125,
            "loss": 3.9115688502788544,
            "approx_kl": 0.004388917219330324,
            "wall_seconds": 865.7892046000052
          },
          {
            "update": 245,
            "environment_steps": 1003520,
            "curriculum": 3,
            "completed_episodes": 2570,
            "new_episodes": 10,
            "new_successes": 2,
            "mean_step_reward": -0.2617466151714325,
            "loss": 14.597019016742706,
            "approx_kl": 0.0039013201276247855,
            "wall_seconds": 869.1413888000243
          },
          {
            "update": 246,
            "environment_steps": 1007616,
            "curriculum": 3,
            "completed_episodes": 2573,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2528824508190155,
            "loss": 5.553580015897751,
            "approx_kl": 0.0033862388045236003,
            "wall_seconds": 872.5173512000183
          },
          {
            "update": 247,
            "environment_steps": 1011712,
            "curriculum": 3,
            "completed_episodes": 2577,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.30240559577941895,
            "loss": 6.003125190734863,
            "approx_kl": 0.008150800240400713,
            "wall_seconds": 875.8771371000039
          },
          {
            "update": 248,
            "environment_steps": 1015808,
            "curriculum": 3,
            "completed_episodes": 2580,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.28444769978523254,
            "loss": 6.1424452513456345,
            "approx_kl": 0.0075404354283818975,
            "wall_seconds": 879.2439752000209
          },
          {
            "update": 249,
            "environment_steps": 1019904,
            "curriculum": 3,
            "completed_episodes": 2582,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3059404194355011,
            "loss": 2.6538319662213326,
            "approx_kl": 0.004792209532752167,
            "wall_seconds": 882.6142082000151
          },
          {
            "update": 250,
            "environment_steps": 1024000,
            "curriculum": 3,
            "completed_episodes": 2585,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2713649868965149,
            "loss": 7.5932421535253525,
            "approx_kl": 0.004835293319047196,
            "wall_seconds": 885.9912681000133
          },
          {
            "update": 251,
            "environment_steps": 1028096,
            "curriculum": 3,
            "completed_episodes": 2587,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.31205418705940247,
            "loss": 2.9302922934293747,
            "approx_kl": 0.00477581177619868,
            "wall_seconds": 889.3290444000158
          },
          {
            "update": 252,
            "environment_steps": 1032192,
            "curriculum": 3,
            "completed_episodes": 2592,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.3427673578262329,
            "loss": 5.22965544462204,
            "approx_kl": 0.004093866289622383,
            "wall_seconds": 892.6644968000182
          },
          {
            "update": 253,
            "environment_steps": 1036288,
            "curriculum": 3,
            "completed_episodes": 2596,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3223310708999634,
            "loss": 4.182222366333008,
            "approx_kl": 0.002184172084525926,
            "wall_seconds": 896.025097900012
          },
          {
            "update": 254,
            "environment_steps": 1040384,
            "curriculum": 3,
            "completed_episodes": 2602,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.2922290563583374,
            "loss": 8.147235810756683,
            "approx_kl": 0.006563251325133024,
            "wall_seconds": 899.367539400002
          },
          {
            "update": 255,
            "environment_steps": 1044480,
            "curriculum": 3,
            "completed_episodes": 2608,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.27027636766433716,
            "loss": 7.105056822299957,
            "approx_kl": 0.006293220678344369,
            "wall_seconds": 902.7284133000067
          },
          {
            "update": 256,
            "environment_steps": 1048576,
            "curriculum": 3,
            "completed_episodes": 2610,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2997456192970276,
            "loss": 2.6428098678588867,
            "approx_kl": 0.004273244961950695,
            "wall_seconds": 906.037229700014
          },
          {
            "update": 257,
            "environment_steps": 1052672,
            "curriculum": 3,
            "completed_episodes": 2614,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2773433029651642,
            "loss": 5.098933279514313,
            "approx_kl": 0.004570631150272675,
            "wall_seconds": 909.3952993999992
          },
          {
            "update": 258,
            "environment_steps": 1056768,
            "curriculum": 3,
            "completed_episodes": 2615,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.2814767062664032,
            "loss": 1.7183973640203476,
            "approx_kl": 0.005621574546239572,
            "wall_seconds": 912.7380899999989
          },
          {
            "update": 259,
            "environment_steps": 1060864,
            "curriculum": 3,
            "completed_episodes": 2620,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2629552483558655,
            "loss": 7.69714942574501,
            "approx_kl": 0.003167388691508677,
            "wall_seconds": 916.1004612000252
          },
          {
            "update": 260,
            "environment_steps": 1064960,
            "curriculum": 3,
            "completed_episodes": 2623,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.292569637298584,
            "loss": 6.2027378231287,
            "approx_kl": 0.003906329482560977,
            "wall_seconds": 919.4545944000129
          },
          {
            "update": 261,
            "environment_steps": 1069056,
            "curriculum": 3,
            "completed_episodes": 2628,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.28808143734931946,
            "loss": 6.2560703456401825,
            "approx_kl": 0.003411025038076332,
            "wall_seconds": 922.8105341000191
          },
          {
            "update": 262,
            "environment_steps": 1073152,
            "curriculum": 3,
            "completed_episodes": 2634,
            "new_episodes": 6,
            "new_successes": 3,
            "mean_step_reward": -0.29231396317481995,
            "loss": 8.652921795845032,
            "approx_kl": 0.004740559805213707,
            "wall_seconds": 926.1550469000067
          },
          {
            "update": 263,
            "environment_steps": 1077248,
            "curriculum": 3,
            "completed_episodes": 2639,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.30748558044433594,
            "loss": 5.32677161693573,
            "approx_kl": 0.0053294377685233485,
            "wall_seconds": 929.4911568000098
          },
          {
            "update": 264,
            "environment_steps": 1081344,
            "curriculum": 3,
            "completed_episodes": 2646,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2579309046268463,
            "loss": 12.104524791240692,
            "approx_kl": 0.004676301865401911,
            "wall_seconds": 932.8772301000135
          },
          {
            "update": 265,
            "environment_steps": 1085440,
            "curriculum": 3,
            "completed_episodes": 2653,
            "new_episodes": 7,
            "new_successes": 4,
            "mean_step_reward": -0.21988323330879211,
            "loss": 14.593144357204437,
            "approx_kl": 0.005686526325007435,
            "wall_seconds": 936.255927900027
          },
          {
            "update": 266,
            "environment_steps": 1089536,
            "curriculum": 3,
            "completed_episodes": 2656,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.23125359416007996,
            "loss": 6.336117833852768,
            "approx_kl": 0.003563936938007828,
            "wall_seconds": 939.6273832000152
          },
          {
            "update": 267,
            "environment_steps": 1093632,
            "curriculum": 3,
            "completed_episodes": 2659,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2649277448654175,
            "loss": 5.1217843145132065,
            "approx_kl": 0.0027618474668997806,
            "wall_seconds": 942.9655299000151
          },
          {
            "update": 268,
            "environment_steps": 1097728,
            "curriculum": 3,
            "completed_episodes": 2661,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2989347577095032,
            "loss": 3.8383112400770187,
            "approx_kl": 0.00452213195967488,
            "wall_seconds": 946.3026757000189
          },
          {
            "update": 269,
            "environment_steps": 1101824,
            "curriculum": 3,
            "completed_episodes": 2669,
            "new_episodes": 8,
            "new_successes": 3,
            "mean_step_reward": -0.295137494802475,
            "loss": 10.6519615650177,
            "approx_kl": 0.0035099055094178766,
            "wall_seconds": 949.638936000003
          },
          {
            "update": 270,
            "environment_steps": 1105920,
            "curriculum": 3,
            "completed_episodes": 2674,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.2670890986919403,
            "loss": 7.329365581274033,
            "approx_kl": 0.003104950530541828,
            "wall_seconds": 952.984767100017
          },
          {
            "update": 271,
            "environment_steps": 1110016,
            "curriculum": 3,
            "completed_episodes": 2677,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.31997591257095337,
            "loss": 3.1319297403097153,
            "approx_kl": 0.004950600432493957,
            "wall_seconds": 956.3316797000007
          },
          {
            "update": 272,
            "environment_steps": 1114112,
            "curriculum": 3,
            "completed_episodes": 2680,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.27962931990623474,
            "loss": 4.8777715265750885,
            "approx_kl": 0.002570974345871946,
            "wall_seconds": 959.6812275000266
          },
          {
            "update": 273,
            "environment_steps": 1118208,
            "curriculum": 3,
            "completed_episodes": 2685,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.27255240082740784,
            "loss": 6.452922135591507,
            "approx_kl": 0.003502270810713526,
            "wall_seconds": 963.0293511000054
          },
          {
            "update": 274,
            "environment_steps": 1122304,
            "curriculum": 3,
            "completed_episodes": 2689,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2751286029815674,
            "loss": 4.9700014144182205,
            "approx_kl": 0.004346354493463878,
            "wall_seconds": 966.3748997000221
          },
          {
            "update": 275,
            "environment_steps": 1126400,
            "curriculum": 3,
            "completed_episodes": 2694,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.2680201530456543,
            "loss": 5.406399458646774,
            "approx_kl": 0.0038126905674289446,
            "wall_seconds": 969.7231237000087
          },
          {
            "update": 276,
            "environment_steps": 1130496,
            "curriculum": 3,
            "completed_episodes": 2698,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.25427210330963135,
            "loss": 6.801907017827034,
            "approx_kl": 0.004047724160045618,
            "wall_seconds": 973.0407271000149
          },
          {
            "update": 277,
            "environment_steps": 1134592,
            "curriculum": 3,
            "completed_episodes": 2699,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3080332279205322,
            "loss": 2.131032794713974,
            "approx_kl": 0.0027706153705366887,
            "wall_seconds": 976.3706587000052
          },
          {
            "update": 278,
            "environment_steps": 1138688,
            "curriculum": 3,
            "completed_episodes": 2704,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.3020245432853699,
            "loss": 5.893214732408524,
            "approx_kl": 0.0077765935311617795,
            "wall_seconds": 979.7214203000185
          },
          {
            "update": 279,
            "environment_steps": 1142784,
            "curriculum": 3,
            "completed_episodes": 2708,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.29666221141815186,
            "loss": 4.603442996740341,
            "approx_kl": 0.0067576061373983975,
            "wall_seconds": 983.0877933000156
          },
          {
            "update": 280,
            "environment_steps": 1146880,
            "curriculum": 3,
            "completed_episodes": 2712,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.283969521522522,
            "loss": 6.586447298526764,
            "approx_kl": 0.0042739824384625535,
            "wall_seconds": 986.4485737000068
          },
          {
            "update": 281,
            "environment_steps": 1150976,
            "curriculum": 3,
            "completed_episodes": 2717,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.2531617283821106,
            "loss": 5.53650888800621,
            "approx_kl": 0.003599034815124469,
            "wall_seconds": 989.8687543000269
          },
          {
            "update": 282,
            "environment_steps": 1155072,
            "curriculum": 3,
            "completed_episodes": 2723,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.26264771819114685,
            "loss": 8.020563513040543,
            "approx_kl": 0.003575167324015638,
            "wall_seconds": 993.2219872000278
          },
          {
            "update": 283,
            "environment_steps": 1159168,
            "curriculum": 3,
            "completed_episodes": 2727,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.24963462352752686,
            "loss": 6.432813405990601,
            "approx_kl": 0.003706766688992502,
            "wall_seconds": 996.5792650000076
          },
          {
            "update": 284,
            "environment_steps": 1163264,
            "curriculum": 3,
            "completed_episodes": 2731,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.24036911129951477,
            "loss": 5.801415741443634,
            "approx_kl": 0.00506680554462946,
            "wall_seconds": 999.9849964000168
          },
          {
            "update": 285,
            "environment_steps": 1167360,
            "curriculum": 3,
            "completed_episodes": 2737,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.23619666695594788,
            "loss": 9.293962061405182,
            "approx_kl": 0.004143777514400426,
            "wall_seconds": 1003.3544249000261
          },
          {
            "update": 286,
            "environment_steps": 1171456,
            "curriculum": 3,
            "completed_episodes": 2742,
            "new_episodes": 5,
            "new_successes": 4,
            "mean_step_reward": -0.26397505402565,
            "loss": 3.7381283044815063,
            "approx_kl": 0.001722084685752634,
            "wall_seconds": 1006.7108849000069
          },
          {
            "update": 287,
            "environment_steps": 1175552,
            "curriculum": 3,
            "completed_episodes": 2743,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3229575753211975,
            "loss": 1.5712324902415276,
            "approx_kl": 0.0030008482790435664,
            "wall_seconds": 1010.0003756000078
          },
          {
            "update": 288,
            "environment_steps": 1179648,
            "curriculum": 3,
            "completed_episodes": 2749,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.28365612030029297,
            "loss": 7.030512124300003,
            "approx_kl": 0.0055627895817451645,
            "wall_seconds": 1013.3237325000227
          },
          {
            "update": 289,
            "environment_steps": 1183744,
            "curriculum": 3,
            "completed_episodes": 2756,
            "new_episodes": 7,
            "new_successes": 3,
            "mean_step_reward": -0.2582114338874817,
            "loss": 9.6860933303833,
            "approx_kl": 0.005175092021090677,
            "wall_seconds": 1016.7089631000126
          },
          {
            "update": 290,
            "environment_steps": 1187840,
            "curriculum": 3,
            "completed_episodes": 2758,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.23611636459827423,
            "loss": 3.0254745483398438,
            "approx_kl": 0.0018414821170154028,
            "wall_seconds": 1020.0670333000016
          },
          {
            "update": 291,
            "environment_steps": 1191936,
            "curriculum": 3,
            "completed_episodes": 2762,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.287050724029541,
            "loss": 4.278399959206581,
            "approx_kl": 0.0023270151104952674,
            "wall_seconds": 1023.4325133000093
          },
          {
            "update": 292,
            "environment_steps": 1196032,
            "curriculum": 3,
            "completed_episodes": 2764,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30060523748397827,
            "loss": 2.2018291503190994,
            "approx_kl": 0.0045005891552136745,
            "wall_seconds": 1026.782378600008
          },
          {
            "update": 293,
            "environment_steps": 1200128,
            "curriculum": 3,
            "completed_episodes": 2769,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.32649967074394226,
            "loss": 4.944480150938034,
            "approx_kl": 0.004271093428542372,
            "wall_seconds": 1030.1282011000148
          },
          {
            "update": 294,
            "environment_steps": 1204224,
            "curriculum": 3,
            "completed_episodes": 2772,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.30795636773109436,
            "loss": 3.3940944224596024,
            "approx_kl": 0.0058153646932623815,
            "wall_seconds": 1033.4876235000265
          },
          {
            "update": 295,
            "environment_steps": 1208320,
            "curriculum": 3,
            "completed_episodes": 2778,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.264434278011322,
            "loss": 5.5939072370529175,
            "approx_kl": 0.002202738502091961,
            "wall_seconds": 1036.821890000021
          },
          {
            "update": 296,
            "environment_steps": 1212416,
            "curriculum": 3,
            "completed_episodes": 2781,
            "new_episodes": 3,
            "new_successes": 3,
            "mean_step_reward": -0.23752808570861816,
            "loss": 1.5611022487282753,
            "approx_kl": 0.002584148769528838,
            "wall_seconds": 1040.1707494000148
          },
          {
            "update": 297,
            "environment_steps": 1216512,
            "curriculum": 3,
            "completed_episodes": 2787,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2678212523460388,
            "loss": 6.135735660791397,
            "approx_kl": 0.005032092733017635,
            "wall_seconds": 1043.5058449000062
          },
          {
            "update": 298,
            "environment_steps": 1220608,
            "curriculum": 3,
            "completed_episodes": 2793,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.25681573152542114,
            "loss": 6.92779341340065,
            "approx_kl": 0.0033914272898982745,
            "wall_seconds": 1046.8689359000127
          },
          {
            "update": 299,
            "environment_steps": 1224704,
            "curriculum": 3,
            "completed_episodes": 2802,
            "new_episodes": 9,
            "new_successes": 4,
            "mean_step_reward": -0.2550398111343384,
            "loss": 8.953265577554703,
            "approx_kl": 0.003595229110942455,
            "wall_seconds": 1050.2377729
          },
          {
            "update": 300,
            "environment_steps": 1228800,
            "curriculum": 3,
            "completed_episodes": 2805,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.275206983089447,
            "loss": 3.5183031633496284,
            "approx_kl": 0.004945929518726189,
            "wall_seconds": 1053.5910043000185
          },
          {
            "update": 301,
            "environment_steps": 1232896,
            "curriculum": 3,
            "completed_episodes": 2808,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.25602561235427856,
            "loss": 4.9410577565431595,
            "approx_kl": 0.002238973909697961,
            "wall_seconds": 1056.9596556000179
          },
          {
            "update": 302,
            "environment_steps": 1236992,
            "curriculum": 3,
            "completed_episodes": 2811,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.3043595552444458,
            "loss": 2.588238440454006,
            "approx_kl": 0.0022661506263830233,
            "wall_seconds": 1060.2851098000247
          },
          {
            "update": 303,
            "environment_steps": 1241088,
            "curriculum": 3,
            "completed_episodes": 2815,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.26478633284568787,
            "loss": 5.780474603176117,
            "approx_kl": 0.003773837881453801,
            "wall_seconds": 1063.6517055000004
          },
          {
            "update": 304,
            "environment_steps": 1245184,
            "curriculum": 3,
            "completed_episodes": 2817,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30299925804138184,
            "loss": 3.0587692707777023,
            "approx_kl": 0.003663652765681036,
            "wall_seconds": 1066.9810854000098
          },
          {
            "update": 305,
            "environment_steps": 1249280,
            "curriculum": 3,
            "completed_episodes": 2823,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.3142799437046051,
            "loss": 5.733986079692841,
            "approx_kl": 0.006137157917692093,
            "wall_seconds": 1070.3060609000095
          },
          {
            "update": 306,
            "environment_steps": 1253376,
            "curriculum": 3,
            "completed_episodes": 2825,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2936934232711792,
            "loss": 3.1104679480195045,
            "approx_kl": 0.004231261144013843,
            "wall_seconds": 1073.6514054000145
          },
          {
            "update": 307,
            "environment_steps": 1257472,
            "curriculum": 3,
            "completed_episodes": 2830,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2937759757041931,
            "loss": 5.4156650602817535,
            "approx_kl": 0.0029052755162410904,
            "wall_seconds": 1076.9872684000002
          },
          {
            "update": 308,
            "environment_steps": 1261568,
            "curriculum": 3,
            "completed_episodes": 2836,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2813212275505066,
            "loss": 7.67840912938118,
            "approx_kl": 0.003319215957162669,
            "wall_seconds": 1080.3269687000138
          },
          {
            "update": 309,
            "environment_steps": 1265664,
            "curriculum": 3,
            "completed_episodes": 2842,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.282117635011673,
            "loss": 8.285050868988037,
            "approx_kl": 0.003487213900370989,
            "wall_seconds": 1083.669638899999
          },
          {
            "update": 310,
            "environment_steps": 1269760,
            "curriculum": 3,
            "completed_episodes": 2843,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.2777522802352905,
            "loss": 2.7007350623607635,
            "approx_kl": 0.0028937001297890674,
            "wall_seconds": 1087.018407000025
          },
          {
            "update": 311,
            "environment_steps": 1273856,
            "curriculum": 3,
            "completed_episodes": 2846,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.30418387055397034,
            "loss": 4.9278704226017,
            "approx_kl": 0.004017854924313724,
            "wall_seconds": 1090.3956435
          },
          {
            "update": 312,
            "environment_steps": 1277952,
            "curriculum": 3,
            "completed_episodes": 2850,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2903658449649811,
            "loss": 3.7120093256235123,
            "approx_kl": 0.0024811126349959522,
            "wall_seconds": 1093.7147139000008
          },
          {
            "update": 313,
            "environment_steps": 1282048,
            "curriculum": 3,
            "completed_episodes": 2853,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2820180356502533,
            "loss": 3.2751880884170532,
            "approx_kl": 0.004155210452154279,
            "wall_seconds": 1097.0567089000251
          },
          {
            "update": 314,
            "environment_steps": 1286144,
            "curriculum": 3,
            "completed_episodes": 2859,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.27456915378570557,
            "loss": 8.366728514432907,
            "approx_kl": 0.005003591450076783,
            "wall_seconds": 1100.4157189000107
          },
          {
            "update": 315,
            "environment_steps": 1290240,
            "curriculum": 3,
            "completed_episodes": 2861,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.26738908886909485,
            "loss": 2.9985828548669815,
            "approx_kl": 0.004735843584057875,
            "wall_seconds": 1103.7750414000184
          },
          {
            "update": 316,
            "environment_steps": 1294336,
            "curriculum": 3,
            "completed_episodes": 2864,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.308131605386734,
            "loss": 3.96676205098629,
            "approx_kl": 0.0068623760125774425,
            "wall_seconds": 1107.1246823000256
          },
          {
            "update": 317,
            "environment_steps": 1298432,
            "curriculum": 3,
            "completed_episodes": 2869,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.3102748990058899,
            "loss": 5.224976792931557,
            "approx_kl": 0.0070376436597143766,
            "wall_seconds": 1110.4785246000101
          },
          {
            "update": 318,
            "environment_steps": 1302528,
            "curriculum": 3,
            "completed_episodes": 2876,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2732689678668976,
            "loss": 6.545537292957306,
            "approx_kl": 0.003289358806796372,
            "wall_seconds": 1113.8385190000117
          },
          {
            "update": 319,
            "environment_steps": 1306624,
            "curriculum": 3,
            "completed_episodes": 2878,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.28350576758384705,
            "loss": 3.2181045338511467,
            "approx_kl": 0.0058349565078970045,
            "wall_seconds": 1117.2114997000026
          },
          {
            "update": 320,
            "environment_steps": 1310720,
            "curriculum": 3,
            "completed_episodes": 2880,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2756742835044861,
            "loss": 2.316736973822117,
            "approx_kl": 0.0040866412564355414,
            "wall_seconds": 1120.850295900018
          },
          {
            "update": 321,
            "environment_steps": 1314816,
            "curriculum": 3,
            "completed_episodes": 2884,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.32090359926223755,
            "loss": 4.25470931828022,
            "approx_kl": 0.00577225639790413,
            "wall_seconds": 1127.2704743000213
          },
          {
            "update": 322,
            "environment_steps": 1318912,
            "curriculum": 3,
            "completed_episodes": 2889,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.30675894021987915,
            "loss": 6.912621721625328,
            "approx_kl": 0.0022838895456516184,
            "wall_seconds": 1133.8957102000131
          },
          {
            "update": 323,
            "environment_steps": 1323008,
            "curriculum": 3,
            "completed_episodes": 2895,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.305240660905838,
            "loss": 6.646127164363861,
            "approx_kl": 0.004397522992803715,
            "wall_seconds": 1140.552058100002
          },
          {
            "update": 324,
            "environment_steps": 1327104,
            "curriculum": 3,
            "completed_episodes": 2898,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.3147745132446289,
            "loss": 4.336021572351456,
            "approx_kl": 0.004170852120296331,
            "wall_seconds": 1147.2755312000227
          },
          {
            "update": 325,
            "environment_steps": 1331200,
            "curriculum": 3,
            "completed_episodes": 2900,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.31449025869369507,
            "loss": 2.9915549606084824,
            "approx_kl": 0.004292155703296885,
            "wall_seconds": 1153.960830600001
          },
          {
            "update": 326,
            "environment_steps": 1335296,
            "curriculum": 3,
            "completed_episodes": 2902,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3086276054382324,
            "loss": 3.232458859682083,
            "approx_kl": 0.004967985481926007,
            "wall_seconds": 1160.6275881000038
          },
          {
            "update": 327,
            "environment_steps": 1339392,
            "curriculum": 3,
            "completed_episodes": 2908,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.2931246757507324,
            "loss": 6.275444895029068,
            "approx_kl": 0.003967934411775786,
            "wall_seconds": 1168.8130113000225
          },
          {
            "update": 328,
            "environment_steps": 1343488,
            "curriculum": 3,
            "completed_episodes": 2911,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.26067468523979187,
            "loss": 3.463056519627571,
            "approx_kl": 0.0037807780790899415,
            "wall_seconds": 1177.2545727000106
          },
          {
            "update": 329,
            "environment_steps": 1347584,
            "curriculum": 3,
            "completed_episodes": 2914,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3079099953174591,
            "loss": 3.904401019215584,
            "approx_kl": 0.005023488156439271,
            "wall_seconds": 1183.3456630000146
          },
          {
            "update": 330,
            "environment_steps": 1351680,
            "curriculum": 3,
            "completed_episodes": 2916,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.29211533069610596,
            "loss": 2.7499556690454483,
            "approx_kl": 0.005777026883151848,
            "wall_seconds": 1189.4724271000014
          },
          {
            "update": 331,
            "environment_steps": 1355776,
            "curriculum": 3,
            "completed_episodes": 2920,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.29862409830093384,
            "loss": 5.838503792881966,
            "approx_kl": 0.007681908125960035,
            "wall_seconds": 1195.6830592000042
          },
          {
            "update": 332,
            "environment_steps": 1359872,
            "curriculum": 3,
            "completed_episodes": 2924,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2847866415977478,
            "loss": 3.7972636371850967,
            "approx_kl": 0.006032629709807225,
            "wall_seconds": 1201.786583900015
          },
          {
            "update": 333,
            "environment_steps": 1363968,
            "curriculum": 3,
            "completed_episodes": 2931,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.2869553565979004,
            "loss": 7.324123591184616,
            "approx_kl": 0.004631735471775755,
            "wall_seconds": 1207.9521272000275
          },
          {
            "update": 334,
            "environment_steps": 1368064,
            "curriculum": 3,
            "completed_episodes": 2934,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.29647237062454224,
            "loss": 2.9648911729454994,
            "approx_kl": 0.005582637153565884,
            "wall_seconds": 1214.008854900021
          },
          {
            "update": 335,
            "environment_steps": 1372160,
            "curriculum": 3,
            "completed_episodes": 2937,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.28410521149635315,
            "loss": 2.7570918574929237,
            "approx_kl": 0.004541939262708183,
            "wall_seconds": 1220.1227885000117
          },
          {
            "update": 336,
            "environment_steps": 1376256,
            "curriculum": 3,
            "completed_episodes": 2942,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2858893871307373,
            "loss": 6.279308944940567,
            "approx_kl": 0.0045527638467319775,
            "wall_seconds": 1226.3158433000208
          },
          {
            "update": 337,
            "environment_steps": 1380352,
            "curriculum": 3,
            "completed_episodes": 2946,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2930009365081787,
            "loss": 4.417528882622719,
            "approx_kl": 0.004794061718712328,
            "wall_seconds": 1232.4698585000006
          },
          {
            "update": 338,
            "environment_steps": 1384448,
            "curriculum": 3,
            "completed_episodes": 2948,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.29520756006240845,
            "loss": 1.5688683204352856,
            "approx_kl": 0.004406203002872644,
            "wall_seconds": 1238.567710400006
          },
          {
            "update": 339,
            "environment_steps": 1388544,
            "curriculum": 3,
            "completed_episodes": 2952,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2991780638694763,
            "loss": 2.4520238116383553,
            "approx_kl": 0.007058507635520073,
            "wall_seconds": 1244.7227287999995
          },
          {
            "update": 340,
            "environment_steps": 1392640,
            "curriculum": 3,
            "completed_episodes": 2957,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2859329581260681,
            "loss": 5.560760051012039,
            "approx_kl": 0.006218638911377639,
            "wall_seconds": 1250.9229320000159
          },
          {
            "update": 341,
            "environment_steps": 1396736,
            "curriculum": 3,
            "completed_episodes": 2959,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2942938506603241,
            "loss": 1.4381194300949574,
            "approx_kl": 0.0025578245913493447,
            "wall_seconds": 1257.0944756000245
          },
          {
            "update": 342,
            "environment_steps": 1400832,
            "curriculum": 3,
            "completed_episodes": 2966,
            "new_episodes": 7,
            "new_successes": 0,
            "mean_step_reward": -0.3256538212299347,
            "loss": 7.741118013858795,
            "approx_kl": 0.007577869713713881,
            "wall_seconds": 1262.8468630000134
          },
          {
            "update": 343,
            "environment_steps": 1404928,
            "curriculum": 3,
            "completed_episodes": 2971,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2983924150466919,
            "loss": 5.417112857103348,
            "approx_kl": 0.002210906535765389,
            "wall_seconds": 1268.4415355000237
          },
          {
            "update": 344,
            "environment_steps": 1409024,
            "curriculum": 3,
            "completed_episodes": 2975,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.30758288502693176,
            "loss": 3.3517228066921234,
            "approx_kl": 0.003186497764545493,
            "wall_seconds": 1273.9533813000016
          },
          {
            "update": 345,
            "environment_steps": 1413120,
            "curriculum": 3,
            "completed_episodes": 2979,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2782175540924072,
            "loss": 5.546174168586731,
            "approx_kl": 0.0038297673308989033,
            "wall_seconds": 1279.5973132000072
          },
          {
            "update": 346,
            "environment_steps": 1417216,
            "curriculum": 3,
            "completed_episodes": 2984,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.26160624623298645,
            "loss": 8.835309714078903,
            "approx_kl": 0.0030180005232978147,
            "wall_seconds": 1285.1669467000174
          },
          {
            "update": 347,
            "environment_steps": 1421312,
            "curriculum": 3,
            "completed_episodes": 2986,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.29784107208251953,
            "loss": 2.571471370756626,
            "approx_kl": 0.005740402040828485,
            "wall_seconds": 1290.6671788000094
          },
          {
            "update": 348,
            "environment_steps": 1425408,
            "curriculum": 3,
            "completed_episodes": 2989,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.29058143496513367,
            "loss": 4.516328975558281,
            "approx_kl": 0.004389387624541996,
            "wall_seconds": 1297.497590700019
          },
          {
            "update": 349,
            "environment_steps": 1429504,
            "curriculum": 3,
            "completed_episodes": 2993,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3046501874923706,
            "loss": 5.466647177934647,
            "approx_kl": 0.003300133324955823,
            "wall_seconds": 1305.12877380001
          },
          {
            "update": 350,
            "environment_steps": 1433600,
            "curriculum": 3,
            "completed_episodes": 2996,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29054954648017883,
            "loss": 3.9461988657712936,
            "approx_kl": 0.00373310414943262,
            "wall_seconds": 1312.7421165000123
          },
          {
            "update": 351,
            "environment_steps": 1437696,
            "curriculum": 3,
            "completed_episodes": 3002,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.29841122031211853,
            "loss": 7.83972641825676,
            "approx_kl": 0.0035975561731902417,
            "wall_seconds": 1320.4449823000177
          },
          {
            "update": 352,
            "environment_steps": 1441792,
            "curriculum": 3,
            "completed_episodes": 3007,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.30882495641708374,
            "loss": 5.218322709202766,
            "approx_kl": 0.00226921439752914,
            "wall_seconds": 1328.0900425
          },
          {
            "update": 353,
            "environment_steps": 1445888,
            "curriculum": 3,
            "completed_episodes": 3010,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2820615768432617,
            "loss": 3.9723228216171265,
            "approx_kl": 0.005380493512348039,
            "wall_seconds": 1335.687855600001
          },
          {
            "update": 354,
            "environment_steps": 1449984,
            "curriculum": 3,
            "completed_episodes": 3016,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.26076316833496094,
            "loss": 6.790798366069794,
            "approx_kl": 0.004787326684891013,
            "wall_seconds": 1343.335083300015
          },
          {
            "update": 355,
            "environment_steps": 1454080,
            "curriculum": 3,
            "completed_episodes": 3020,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2518671154975891,
            "loss": 8.25504395365715,
            "approx_kl": 0.004268363914889051,
            "wall_seconds": 1350.090823400009
          },
          {
            "update": 356,
            "environment_steps": 1458176,
            "curriculum": 3,
            "completed_episodes": 3023,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2569287121295929,
            "loss": 4.026709288358688,
            "approx_kl": 0.0054531708992726635,
            "wall_seconds": 1358.415027800016
          },
          {
            "update": 357,
            "environment_steps": 1462272,
            "curriculum": 3,
            "completed_episodes": 3023,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.27228525280952454,
            "loss": 1.1518404334783554,
            "approx_kl": 0.007210072690213565,
            "wall_seconds": 1367.505696200009
          },
          {
            "update": 358,
            "environment_steps": 1466368,
            "curriculum": 3,
            "completed_episodes": 3029,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.30592456459999084,
            "loss": 5.666232317686081,
            "approx_kl": 0.00527732420960092,
            "wall_seconds": 1376.6706727000128
          },
          {
            "update": 359,
            "environment_steps": 1470464,
            "curriculum": 3,
            "completed_episodes": 3034,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.275570273399353,
            "loss": 6.920988917350769,
            "approx_kl": 0.006516290206491249,
            "wall_seconds": 1385.8197771000268
          },
          {
            "update": 360,
            "environment_steps": 1474560,
            "curriculum": 3,
            "completed_episodes": 3038,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.26435387134552,
            "loss": 4.239337310194969,
            "approx_kl": 0.0061274639738257974,
            "wall_seconds": 1395.009950800013
          },
          {
            "update": 361,
            "environment_steps": 1478656,
            "curriculum": 3,
            "completed_episodes": 3044,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.29730913043022156,
            "loss": 6.417852133512497,
            "approx_kl": 0.0024313818976224866,
            "wall_seconds": 1404.216573200014
          },
          {
            "update": 362,
            "environment_steps": 1482752,
            "curriculum": 3,
            "completed_episodes": 3048,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.29458653926849365,
            "loss": 4.560172736644745,
            "approx_kl": 0.004185208657872863,
            "wall_seconds": 1413.3611992000078
          },
          {
            "update": 363,
            "environment_steps": 1486848,
            "curriculum": 3,
            "completed_episodes": 3054,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.27088356018066406,
            "loss": 7.049025058746338,
            "approx_kl": 0.0019777244415308814,
            "wall_seconds": 1421.6207440000144
          },
          {
            "update": 364,
            "environment_steps": 1490944,
            "curriculum": 3,
            "completed_episodes": 3057,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2831551432609558,
            "loss": 2.9400039687752724,
            "approx_kl": 0.002882816032069968,
            "wall_seconds": 1429.2037810000184
          },
          {
            "update": 365,
            "environment_steps": 1495040,
            "curriculum": 3,
            "completed_episodes": 3061,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2907859683036804,
            "loss": 3.8390293568372726,
            "approx_kl": 0.002037262325757183,
            "wall_seconds": 1438.2574960000056
          },
          {
            "update": 366,
            "environment_steps": 1499136,
            "curriculum": 3,
            "completed_episodes": 3062,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.32658523321151733,
            "loss": 1.3639107048511505,
            "approx_kl": 0.0028560099090100266,
            "wall_seconds": 1447.2361587000196
          },
          {
            "update": 367,
            "environment_steps": 1503232,
            "curriculum": 3,
            "completed_episodes": 3065,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2996663451194763,
            "loss": 2.854779563844204,
            "approx_kl": 0.0034177894922322594,
            "wall_seconds": 1456.3366375000041
          },
          {
            "update": 368,
            "environment_steps": 1507328,
            "curriculum": 3,
            "completed_episodes": 3071,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.31018346548080444,
            "loss": 6.638537794351578,
            "approx_kl": 0.005080347833427368,
            "wall_seconds": 1465.538434200018
          },
          {
            "update": 369,
            "environment_steps": 1511424,
            "curriculum": 3,
            "completed_episodes": 3075,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2868695855140686,
            "loss": 4.212557211518288,
            "approx_kl": 0.004100613808986964,
            "wall_seconds": 1474.6646429000248
          },
          {
            "update": 370,
            "environment_steps": 1515520,
            "curriculum": 3,
            "completed_episodes": 3079,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.26485759019851685,
            "loss": 4.505238234996796,
            "approx_kl": 0.006669032634817995,
            "wall_seconds": 1483.8416580000194
          },
          {
            "update": 371,
            "environment_steps": 1519616,
            "curriculum": 3,
            "completed_episodes": 3081,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30838266015052795,
            "loss": 2.7405403405427933,
            "approx_kl": 0.017259526706766337,
            "wall_seconds": 1492.8347166000167
          },
          {
            "update": 372,
            "environment_steps": 1523712,
            "curriculum": 3,
            "completed_episodes": 3085,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3317965865135193,
            "loss": 4.05600506067276,
            "approx_kl": 0.023646985573577695,
            "wall_seconds": 1499.468297400017
          },
          {
            "update": 373,
            "environment_steps": 1527808,
            "curriculum": 3,
            "completed_episodes": 3089,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.32178524136543274,
            "loss": 4.331390991806984,
            "approx_kl": 0.006165463506476954,
            "wall_seconds": 1508.421592700004
          },
          {
            "update": 374,
            "environment_steps": 1531904,
            "curriculum": 3,
            "completed_episodes": 3092,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2792838215827942,
            "loss": 2.7042689695954323,
            "approx_kl": 0.005120821013406385,
            "wall_seconds": 1517.5442942000227
          },
          {
            "update": 375,
            "environment_steps": 1536000,
            "curriculum": 3,
            "completed_episodes": 3094,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.25187554955482483,
            "loss": 3.0708551928400993,
            "approx_kl": 0.005218851430981886,
            "wall_seconds": 1526.6340212000068
          },
          {
            "update": 376,
            "environment_steps": 1540096,
            "curriculum": 3,
            "completed_episodes": 3096,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.27922630310058594,
            "loss": 2.8845290765166283,
            "approx_kl": 0.0067001684292335995,
            "wall_seconds": 1535.8010895000189
          },
          {
            "update": 377,
            "environment_steps": 1544192,
            "curriculum": 3,
            "completed_episodes": 3101,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.27403461933135986,
            "loss": 5.302853032946587,
            "approx_kl": 0.00648229798025568,
            "wall_seconds": 1545.0434251000115
          },
          {
            "update": 378,
            "environment_steps": 1548288,
            "curriculum": 3,
            "completed_episodes": 3105,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3079938590526581,
            "loss": 5.161706775426865,
            "approx_kl": 0.008172008969268063,
            "wall_seconds": 1554.2092827000015
          },
          {
            "update": 379,
            "environment_steps": 1552384,
            "curriculum": 3,
            "completed_episodes": 3108,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.3058725893497467,
            "loss": 3.508385255932808,
            "approx_kl": 0.011950614458328346,
            "wall_seconds": 1563.3161724000238
          },
          {
            "update": 380,
            "environment_steps": 1556480,
            "curriculum": 3,
            "completed_episodes": 3113,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.3232015371322632,
            "loss": 5.119090348482132,
            "approx_kl": 0.009181707951938733,
            "wall_seconds": 1569.7696030000225
          },
          {
            "update": 381,
            "environment_steps": 1560576,
            "curriculum": 3,
            "completed_episodes": 3115,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3016054034233093,
            "loss": 2.4154164120554924,
            "approx_kl": 0.006368238984578056,
            "wall_seconds": 1577.8595634000085
          },
          {
            "update": 382,
            "environment_steps": 1564672,
            "curriculum": 3,
            "completed_episodes": 3120,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.27534016966819763,
            "loss": 8.27464410662651,
            "approx_kl": 0.005090602327982197,
            "wall_seconds": 1587.075256800017
          },
          {
            "update": 383,
            "environment_steps": 1568768,
            "curriculum": 3,
            "completed_episodes": 3125,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.27244874835014343,
            "loss": 5.387698829174042,
            "approx_kl": 0.004259137327608187,
            "wall_seconds": 1596.2493734000018
          },
          {
            "update": 384,
            "environment_steps": 1572864,
            "curriculum": 3,
            "completed_episodes": 3130,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.25744861364364624,
            "loss": 7.289171785116196,
            "approx_kl": 0.0031544505327474326,
            "wall_seconds": 1605.4555132000241
          },
          {
            "update": 385,
            "environment_steps": 1576960,
            "curriculum": 3,
            "completed_episodes": 3132,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.27506881952285767,
            "loss": 2.6736153215169907,
            "approx_kl": 0.0028901941514050122,
            "wall_seconds": 1614.6457472000038
          },
          {
            "update": 386,
            "environment_steps": 1581056,
            "curriculum": 3,
            "completed_episodes": 3137,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.29697585105895996,
            "loss": 7.257439583539963,
            "approx_kl": 0.0048475842813786585,
            "wall_seconds": 1623.1116180000245
          },
          {
            "update": 387,
            "environment_steps": 1585152,
            "curriculum": 3,
            "completed_episodes": 3142,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.24263226985931396,
            "loss": 3.528351679444313,
            "approx_kl": 0.004514376174483914,
            "wall_seconds": 1630.674491600017
          },
          {
            "update": 388,
            "environment_steps": 1589248,
            "curriculum": 3,
            "completed_episodes": 3147,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2890852689743042,
            "loss": 5.135756000876427,
            "approx_kl": 0.004402731039590435,
            "wall_seconds": 1639.75646840001
          },
          {
            "update": 389,
            "environment_steps": 1593344,
            "curriculum": 3,
            "completed_episodes": 3151,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.295292466878891,
            "loss": 5.4573129415512085,
            "approx_kl": 0.0026541190563875716,
            "wall_seconds": 1648.8833058000018
          },
          {
            "update": 390,
            "environment_steps": 1597440,
            "curriculum": 3,
            "completed_episodes": 3151,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.3004078269004822,
            "loss": 0.5982703436166048,
            "approx_kl": 0.0052115345461061224,
            "wall_seconds": 1657.982580300013
          },
          {
            "update": 391,
            "environment_steps": 1601536,
            "curriculum": 3,
            "completed_episodes": 3156,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2745392918586731,
            "loss": 5.832456201314926,
            "approx_kl": 0.005700163372239331,
            "wall_seconds": 1667.1883166000189
          },
          {
            "update": 392,
            "environment_steps": 1605632,
            "curriculum": 3,
            "completed_episodes": 3162,
            "new_episodes": 6,
            "new_successes": 3,
            "mean_step_reward": -0.2756385803222656,
            "loss": 6.408386260271072,
            "approx_kl": 0.0024764080008026212,
            "wall_seconds": 1676.357238800003
          },
          {
            "update": 393,
            "environment_steps": 1609728,
            "curriculum": 3,
            "completed_episodes": 3165,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29051464796066284,
            "loss": 4.266441911458969,
            "approx_kl": 0.002970342269691173,
            "wall_seconds": 1685.5591440999997
          },
          {
            "update": 394,
            "environment_steps": 1613824,
            "curriculum": 3,
            "completed_episodes": 3171,
            "new_episodes": 6,
            "new_successes": 4,
            "mean_step_reward": -0.2733803987503052,
            "loss": 9.040444016456604,
            "approx_kl": 0.0056759348772175144,
            "wall_seconds": 1694.629594900005
          },
          {
            "update": 395,
            "environment_steps": 1617920,
            "curriculum": 3,
            "completed_episodes": 3175,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3044207692146301,
            "loss": 4.089529871940613,
            "approx_kl": 0.004094931500731036,
            "wall_seconds": 1701.482589000021
          },
          {
            "update": 396,
            "environment_steps": 1622016,
            "curriculum": 3,
            "completed_episodes": 3180,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2479802519083023,
            "loss": 7.490034580230713,
            "approx_kl": 0.0035356695734662935,
            "wall_seconds": 1710.5636401000083
          },
          {
            "update": 397,
            "environment_steps": 1626112,
            "curriculum": 3,
            "completed_episodes": 3187,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.27877044677734375,
            "loss": 6.412697613239288,
            "approx_kl": 0.0028293614013819024,
            "wall_seconds": 1719.6924839000276
          },
          {
            "update": 398,
            "environment_steps": 1630208,
            "curriculum": 3,
            "completed_episodes": 3191,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.26308608055114746,
            "loss": 4.745972841978073,
            "approx_kl": 0.0027913355188502464,
            "wall_seconds": 1728.8688308000274
          },
          {
            "update": 399,
            "environment_steps": 1634304,
            "curriculum": 3,
            "completed_episodes": 3193,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2699447274208069,
            "loss": 2.874777764081955,
            "approx_kl": 0.002038488604739541,
            "wall_seconds": 1738.0949763000244
          },
          {
            "update": 400,
            "environment_steps": 1638400,
            "curriculum": 3,
            "completed_episodes": 3196,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3088953495025635,
            "loss": 3.4543007761240005,
            "approx_kl": 0.00337444581600721,
            "wall_seconds": 1747.232292700006
          },
          {
            "update": 401,
            "environment_steps": 1642496,
            "curriculum": 3,
            "completed_episodes": 3199,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3059163987636566,
            "loss": 3.7919194400310516,
            "approx_kl": 0.005049176590546267,
            "wall_seconds": 1756.4279725000088
          },
          {
            "update": 402,
            "environment_steps": 1646592,
            "curriculum": 3,
            "completed_episodes": 3202,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3031049966812134,
            "loss": 4.230462357401848,
            "approx_kl": 0.003317961371067213,
            "wall_seconds": 1765.6940537000191
          },
          {
            "update": 403,
            "environment_steps": 1650688,
            "curriculum": 3,
            "completed_episodes": 3205,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3285321593284607,
            "loss": 3.04183891415596,
            "approx_kl": 0.005998431628540857,
            "wall_seconds": 1772.5007308000058
          },
          {
            "update": 404,
            "environment_steps": 1654784,
            "curriculum": 3,
            "completed_episodes": 3207,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3114769458770752,
            "loss": 2.752630203962326,
            "approx_kl": 0.0027061972687079106,
            "wall_seconds": 1781.452436600026
          },
          {
            "update": 405,
            "environment_steps": 1658880,
            "curriculum": 3,
            "completed_episodes": 3213,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.28999757766723633,
            "loss": 5.885021984577179,
            "approx_kl": 0.0025027996125572827,
            "wall_seconds": 1790.5473951000022
          },
          {
            "update": 406,
            "environment_steps": 1662976,
            "curriculum": 3,
            "completed_episodes": 3219,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.29109513759613037,
            "loss": 6.371092230081558,
            "approx_kl": 0.004543768209259724,
            "wall_seconds": 1799.690044500021
          },
          {
            "update": 407,
            "environment_steps": 1667072,
            "curriculum": 3,
            "completed_episodes": 3223,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.27944880723953247,
            "loss": 5.103715658187866,
            "approx_kl": 0.0025985512002080213,
            "wall_seconds": 1808.9885414000018
          },
          {
            "update": 408,
            "environment_steps": 1671168,
            "curriculum": 3,
            "completed_episodes": 3227,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.29413527250289917,
            "loss": 5.3321203738451,
            "approx_kl": 0.005560248777328525,
            "wall_seconds": 1818.5529444000276
          },
          {
            "update": 409,
            "environment_steps": 1675264,
            "curriculum": 3,
            "completed_episodes": 3229,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2937796115875244,
            "loss": 2.3827858343720436,
            "approx_kl": 0.004278730175428791,
            "wall_seconds": 1827.7498615000222
          },
          {
            "update": 410,
            "environment_steps": 1679360,
            "curriculum": 3,
            "completed_episodes": 3233,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.30216705799102783,
            "loss": 5.675070911645889,
            "approx_kl": 0.0031333017141150776,
            "wall_seconds": 1837.0043858000136
          },
          {
            "update": 411,
            "environment_steps": 1683456,
            "curriculum": 3,
            "completed_episodes": 3238,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.3035503625869751,
            "loss": 5.329130485653877,
            "approx_kl": 0.0036233795508451294,
            "wall_seconds": 1844.1793515000027
          },
          {
            "update": 412,
            "environment_steps": 1687552,
            "curriculum": 3,
            "completed_episodes": 3241,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3001629710197449,
            "loss": 3.7545228004455566,
            "approx_kl": 0.004844522194616729,
            "wall_seconds": 1850.1520453000267
          },
          {
            "update": 413,
            "environment_steps": 1691648,
            "curriculum": 3,
            "completed_episodes": 3243,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.27794349193573,
            "loss": 3.5393918454647064,
            "approx_kl": 0.0063190601467795204,
            "wall_seconds": 1858.0613235000055
          },
          {
            "update": 414,
            "environment_steps": 1695744,
            "curriculum": 3,
            "completed_episodes": 3247,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.28196537494659424,
            "loss": 4.170056566596031,
            "approx_kl": 0.004109948753466597,
            "wall_seconds": 1865.8427873000037
          },
          {
            "update": 415,
            "environment_steps": 1699840,
            "curriculum": 3,
            "completed_episodes": 3252,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.29572534561157227,
            "loss": 5.700524657964706,
            "approx_kl": 0.00467273491085507,
            "wall_seconds": 1873.747880400013
          },
          {
            "update": 416,
            "environment_steps": 1703936,
            "curriculum": 3,
            "completed_episodes": 3258,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.31229132413864136,
            "loss": 6.357669726014137,
            "approx_kl": 0.0022919193579582497,
            "wall_seconds": 1881.601117800019
          },
          {
            "update": 417,
            "environment_steps": 1708032,
            "curriculum": 3,
            "completed_episodes": 3260,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2885815501213074,
            "loss": 2.2371187061071396,
            "approx_kl": 0.004857385058130603,
            "wall_seconds": 1889.4688082000066
          },
          {
            "update": 418,
            "environment_steps": 1712128,
            "curriculum": 3,
            "completed_episodes": 3264,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3044503331184387,
            "loss": 6.996781170368195,
            "approx_kl": 0.0054647862452839036,
            "wall_seconds": 1897.1278605000116
          },
          {
            "update": 419,
            "environment_steps": 1716224,
            "curriculum": 3,
            "completed_episodes": 3267,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.32721197605133057,
            "loss": 3.0946744680404663,
            "approx_kl": 0.004555418378004106,
            "wall_seconds": 1904.8717174000049
          },
          {
            "update": 420,
            "environment_steps": 1720320,
            "curriculum": 3,
            "completed_episodes": 3272,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.325036883354187,
            "loss": 5.443768128752708,
            "approx_kl": 0.00426531720222556,
            "wall_seconds": 1912.6478105000278
          },
          {
            "update": 421,
            "environment_steps": 1724416,
            "curriculum": 3,
            "completed_episodes": 3274,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.29258090257644653,
            "loss": 3.143963560461998,
            "approx_kl": 0.0038078409706940874,
            "wall_seconds": 1920.4444905000273
          },
          {
            "update": 422,
            "environment_steps": 1728512,
            "curriculum": 3,
            "completed_episodes": 3277,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3040509819984436,
            "loss": 3.384075805544853,
            "approx_kl": 0.005180507247132482,
            "wall_seconds": 1928.2588485000015
          },
          {
            "update": 423,
            "environment_steps": 1732608,
            "curriculum": 3,
            "completed_episodes": 3283,
            "new_episodes": 6,
            "new_successes": 3,
            "mean_step_reward": -0.2695849537849426,
            "loss": 4.49248531460762,
            "approx_kl": 0.0038836718813399784,
            "wall_seconds": 1936.2053046000074
          },
          {
            "update": 424,
            "environment_steps": 1736704,
            "curriculum": 3,
            "completed_episodes": 3289,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.28946077823638916,
            "loss": 5.336673751473427,
            "approx_kl": 0.006419992005248787,
            "wall_seconds": 1944.1154693000135
          },
          {
            "update": 425,
            "environment_steps": 1740800,
            "curriculum": 3,
            "completed_episodes": 3293,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2606061100959778,
            "loss": 5.015542253851891,
            "approx_kl": 0.004622305004886584,
            "wall_seconds": 1952.0321514000243
          },
          {
            "update": 426,
            "environment_steps": 1744896,
            "curriculum": 3,
            "completed_episodes": 3296,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.27300161123275757,
            "loss": 4.837558686733246,
            "approx_kl": 0.005544154035305837,
            "wall_seconds": 1960.016386000003
          },
          {
            "update": 427,
            "environment_steps": 1748992,
            "curriculum": 3,
            "completed_episodes": 3300,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.29509803652763367,
            "loss": 3.546871244907379,
            "approx_kl": 0.005363430012948811,
            "wall_seconds": 1967.984456700011
          },
          {
            "update": 428,
            "environment_steps": 1753088,
            "curriculum": 3,
            "completed_episodes": 3304,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.30928266048431396,
            "loss": 7.1087846755981445,
            "approx_kl": 0.005668915862770518,
            "wall_seconds": 1975.9632599000179
          },
          {
            "update": 429,
            "environment_steps": 1757184,
            "curriculum": 3,
            "completed_episodes": 3311,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.2722494304180145,
            "loss": 12.4494908452034,
            "approx_kl": 0.004390407128084917,
            "wall_seconds": 1983.9043056000082
          },
          {
            "update": 430,
            "environment_steps": 1761280,
            "curriculum": 3,
            "completed_episodes": 3313,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2935284376144409,
            "loss": 2.504360191524029,
            "approx_kl": 0.005113956889545079,
            "wall_seconds": 1991.870886800025
          },
          {
            "update": 431,
            "environment_steps": 1765376,
            "curriculum": 3,
            "completed_episodes": 3316,
            "new_episodes": 3,
            "new_successes": 2,
            "mean_step_reward": -0.2831372618675232,
            "loss": 5.521352380514145,
            "approx_kl": 0.005259121036942815,
            "wall_seconds": 1999.8639138000144
          },
          {
            "update": 432,
            "environment_steps": 1769472,
            "curriculum": 3,
            "completed_episodes": 3319,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.30068182945251465,
            "loss": 3.3901254683732986,
            "approx_kl": 0.006405081530829193,
            "wall_seconds": 2007.8279072000005
          },
          {
            "update": 433,
            "environment_steps": 1773568,
            "curriculum": 3,
            "completed_episodes": 3325,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.2846299409866333,
            "loss": 8.020069688558578,
            "approx_kl": 0.00463194213443785,
            "wall_seconds": 2015.8918427000172
          },
          {
            "update": 434,
            "environment_steps": 1777664,
            "curriculum": 3,
            "completed_episodes": 3331,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2764112651348114,
            "loss": 6.822334140539169,
            "approx_kl": 0.005157894232979743,
            "wall_seconds": 2023.9016428000177
          },
          {
            "update": 435,
            "environment_steps": 1781760,
            "curriculum": 3,
            "completed_episodes": 3337,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.27051985263824463,
            "loss": 6.528808906674385,
            "approx_kl": 0.0019002448607352562,
            "wall_seconds": 2031.890350600006
          },
          {
            "update": 436,
            "environment_steps": 1785856,
            "curriculum": 3,
            "completed_episodes": 3341,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2985544204711914,
            "loss": 6.003951728343964,
            "approx_kl": 0.003184428558597574,
            "wall_seconds": 2039.8610985000269
          },
          {
            "update": 437,
            "environment_steps": 1789952,
            "curriculum": 3,
            "completed_episodes": 3343,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.27077823877334595,
            "loss": 3.003048524260521,
            "approx_kl": 0.003839943354250863,
            "wall_seconds": 2047.7965544000035
          },
          {
            "update": 438,
            "environment_steps": 1794048,
            "curriculum": 3,
            "completed_episodes": 3349,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2682648301124573,
            "loss": 5.805856749415398,
            "approx_kl": 0.004904726916720392,
            "wall_seconds": 2055.809631400014
          },
          {
            "update": 439,
            "environment_steps": 1798144,
            "curriculum": 3,
            "completed_episodes": 3354,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.26498356461524963,
            "loss": 5.78358069062233,
            "approx_kl": 0.0022737568251613993,
            "wall_seconds": 2063.790402600018
          },
          {
            "update": 440,
            "environment_steps": 1802240,
            "curriculum": 3,
            "completed_episodes": 3356,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.2547954320907593,
            "loss": 2.632885344326496,
            "approx_kl": 0.002295242575200973,
            "wall_seconds": 2071.703411100025
          },
          {
            "update": 441,
            "environment_steps": 1806336,
            "curriculum": 3,
            "completed_episodes": 3359,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3277452886104584,
            "loss": 3.0270770341157913,
            "approx_kl": 0.005908966020797379,
            "wall_seconds": 2079.711292000022
          },
          {
            "update": 442,
            "environment_steps": 1810432,
            "curriculum": 3,
            "completed_episodes": 3362,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.32739269733428955,
            "loss": 3.7215067595243454,
            "approx_kl": 0.007279913996171672,
            "wall_seconds": 2087.691171100014
          },
          {
            "update": 443,
            "environment_steps": 1814528,
            "curriculum": 3,
            "completed_episodes": 3369,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.3124738931655884,
            "loss": 6.57119032740593,
            "approx_kl": 0.007130536152544664,
            "wall_seconds": 2095.648129900015
          },
          {
            "update": 444,
            "environment_steps": 1818624,
            "curriculum": 3,
            "completed_episodes": 3376,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.31170880794525146,
            "loss": 4.782641276717186,
            "approx_kl": 0.005190375188249163,
            "wall_seconds": 2103.652653500001
          },
          {
            "update": 445,
            "environment_steps": 1822720,
            "curriculum": 3,
            "completed_episodes": 3379,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.29281291365623474,
            "loss": 2.2698599621653557,
            "approx_kl": 0.008770993572397856,
            "wall_seconds": 2111.6256228000275
          },
          {
            "update": 446,
            "environment_steps": 1826816,
            "curriculum": 3,
            "completed_episodes": 3382,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.29917436838150024,
            "loss": 4.072827726602554,
            "approx_kl": 0.007113389921869384,
            "wall_seconds": 2119.6059695000004
          },
          {
            "update": 447,
            "environment_steps": 1830912,
            "curriculum": 3,
            "completed_episodes": 3386,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2886609435081482,
            "loss": 3.3634813874959946,
            "approx_kl": 0.004623836761311395,
            "wall_seconds": 2125.4312781000044
          },
          {
            "update": 448,
            "environment_steps": 1835008,
            "curriculum": 3,
            "completed_episodes": 3391,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2947601079940796,
            "loss": 7.466098755598068,
            "approx_kl": 0.0031135221979639027,
            "wall_seconds": 2131.5725351000146
          },
          {
            "update": 449,
            "environment_steps": 1839104,
            "curriculum": 3,
            "completed_episodes": 3396,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.3080311417579651,
            "loss": 7.797743558883667,
            "approx_kl": 0.0037439361731230747,
            "wall_seconds": 2140.4234735000064
          },
          {
            "update": 450,
            "environment_steps": 1843200,
            "curriculum": 3,
            "completed_episodes": 3399,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2906810939311981,
            "loss": 7.105254232883453,
            "approx_kl": 0.0038207928264455404,
            "wall_seconds": 2152.5869830000156
          },
          {
            "update": 451,
            "environment_steps": 1847296,
            "curriculum": 3,
            "completed_episodes": 3404,
            "new_episodes": 5,
            "new_successes": 4,
            "mean_step_reward": -0.2746264636516571,
            "loss": 6.298306420445442,
            "approx_kl": 0.0019288054463686422,
            "wall_seconds": 2163.703837600013
          },
          {
            "update": 452,
            "environment_steps": 1851392,
            "curriculum": 3,
            "completed_episodes": 3409,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.31593143939971924,
            "loss": 6.49188107252121,
            "approx_kl": 0.003349550846905913,
            "wall_seconds": 2176.26600850001
          },
          {
            "update": 453,
            "environment_steps": 1855488,
            "curriculum": 3,
            "completed_episodes": 3416,
            "new_episodes": 7,
            "new_successes": 3,
            "mean_step_reward": -0.24783998727798462,
            "loss": 11.508175730705261,
            "approx_kl": 0.0029327089541766327,
            "wall_seconds": 2188.9042565000127
          },
          {
            "update": 454,
            "environment_steps": 1859584,
            "curriculum": 3,
            "completed_episodes": 3423,
            "new_episodes": 7,
            "new_successes": 4,
            "mean_step_reward": -0.26983609795570374,
            "loss": 7.905084848403931,
            "approx_kl": 0.0020791599781659897,
            "wall_seconds": 2201.6950521000254
          },
          {
            "update": 455,
            "environment_steps": 1863680,
            "curriculum": 3,
            "completed_episodes": 3424,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.3142726421356201,
            "loss": 0.8352148532867432,
            "approx_kl": 0.0036533668207994197,
            "wall_seconds": 2214.3987030000135
          },
          {
            "update": 456,
            "environment_steps": 1867776,
            "curriculum": 3,
            "completed_episodes": 3427,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.32214444875717163,
            "loss": 3.441451445221901,
            "approx_kl": 0.004001634570158785,
            "wall_seconds": 2227.1797842000087
          },
          {
            "update": 457,
            "environment_steps": 1871872,
            "curriculum": 3,
            "completed_episodes": 3431,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.31384044885635376,
            "loss": 4.995752647519112,
            "approx_kl": 0.0025302283866039943,
            "wall_seconds": 2238.5311951000185
          },
          {
            "update": 458,
            "environment_steps": 1875968,
            "curriculum": 3,
            "completed_episodes": 3437,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.29582637548446655,
            "loss": 8.037569552659988,
            "approx_kl": 0.003302461507701082,
            "wall_seconds": 2247.2944282000244
          },
          {
            "update": 459,
            "environment_steps": 1880064,
            "curriculum": 3,
            "completed_episodes": 3442,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.28567638993263245,
            "loss": 6.5324854254722595,
            "approx_kl": 0.005439180091343587,
            "wall_seconds": 2254.477525800001
          },
          {
            "update": 460,
            "environment_steps": 1884160,
            "curriculum": 3,
            "completed_episodes": 3447,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.29379475116729736,
            "loss": 4.860760390758514,
            "approx_kl": 0.0029581578783108853,
            "wall_seconds": 2263.533441000007
          },
          {
            "update": 461,
            "environment_steps": 1888256,
            "curriculum": 3,
            "completed_episodes": 3449,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.3189150094985962,
            "loss": 3.1643787547945976,
            "approx_kl": 0.004978496748663019,
            "wall_seconds": 2272.597176600015
          },
          {
            "update": 462,
            "environment_steps": 1892352,
            "curriculum": 3,
            "completed_episodes": 3455,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.31268757581710815,
            "loss": 6.223696261644363,
            "approx_kl": 0.005437434334453428,
            "wall_seconds": 2281.7246619000216
          },
          {
            "update": 463,
            "environment_steps": 1896448,
            "curriculum": 3,
            "completed_episodes": 3458,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3125343918800354,
            "loss": 3.967219203710556,
            "approx_kl": 0.0017468982150603551,
            "wall_seconds": 2290.895094700012
          },
          {
            "update": 464,
            "environment_steps": 1900544,
            "curriculum": 3,
            "completed_episodes": 3462,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.30296790599823,
            "loss": 4.2141420394182205,
            "approx_kl": 0.005526262040802976,
            "wall_seconds": 2300.072532900027
          },
          {
            "update": 465,
            "environment_steps": 1904640,
            "curriculum": 3,
            "completed_episodes": 3464,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.290690541267395,
            "loss": 3.81268772482872,
            "approx_kl": 0.004269662262231577,
            "wall_seconds": 2309.2319527000072
          },
          {
            "update": 466,
            "environment_steps": 1908736,
            "curriculum": 3,
            "completed_episodes": 3466,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.29841718077659607,
            "loss": 3.5343469828367233,
            "approx_kl": 0.0018598116766952444,
            "wall_seconds": 2318.412286400009
          },
          {
            "update": 467,
            "environment_steps": 1912832,
            "curriculum": 3,
            "completed_episodes": 3473,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.2815004885196686,
            "loss": 6.386137098073959,
            "approx_kl": 0.004781104849826079,
            "wall_seconds": 2324.6279910000158
          },
          {
            "update": 468,
            "environment_steps": 1916928,
            "curriculum": 3,
            "completed_episodes": 3478,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.28030717372894287,
            "loss": 8.610895544290543,
            "approx_kl": 0.0023294692182389554,
            "wall_seconds": 2331.8986781999993
          },
          {
            "update": 469,
            "environment_steps": 1921024,
            "curriculum": 3,
            "completed_episodes": 3483,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.3003386855125427,
            "loss": 5.471738949418068,
            "approx_kl": 0.004941285314998822,
            "wall_seconds": 2339.838845100021
          },
          {
            "update": 470,
            "environment_steps": 1925120,
            "curriculum": 3,
            "completed_episodes": 3489,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.29826053977012634,
            "loss": 4.903875917196274,
            "approx_kl": 0.005345958681573393,
            "wall_seconds": 2347.7065550000116
          },
          {
            "update": 471,
            "environment_steps": 1929216,
            "curriculum": 3,
            "completed_episodes": 3493,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.30336254835128784,
            "loss": 4.045656427741051,
            "approx_kl": 0.006560408306540921,
            "wall_seconds": 2355.5485079000064
          },
          {
            "update": 472,
            "environment_steps": 1933312,
            "curriculum": 3,
            "completed_episodes": 3496,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.3041483461856842,
            "loss": 1.8483128547668457,
            "approx_kl": 0.0033338759531034157,
            "wall_seconds": 2363.2708058000135
          },
          {
            "update": 473,
            "environment_steps": 1937408,
            "curriculum": 3,
            "completed_episodes": 3501,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.29850929975509644,
            "loss": 4.530452102422714,
            "approx_kl": 0.004587174091284396,
            "wall_seconds": 2371.0197215000226
          },
          {
            "update": 474,
            "environment_steps": 1941504,
            "curriculum": 3,
            "completed_episodes": 3504,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.3077658712863922,
            "loss": 2.442728564143181,
            "approx_kl": 0.003391454360098578,
            "wall_seconds": 2378.766154400015
          },
          {
            "update": 475,
            "environment_steps": 1945600,
            "curriculum": 3,
            "completed_episodes": 3505,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.33099839091300964,
            "loss": 1.407539315521717,
            "approx_kl": 0.007313952402910218,
            "wall_seconds": 2386.507211600023
          },
          {
            "update": 476,
            "environment_steps": 1949696,
            "curriculum": 3,
            "completed_episodes": 3508,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.328514039516449,
            "loss": 3.5356530845165253,
            "approx_kl": 0.00570063177292468,
            "wall_seconds": 2394.264975099999
          },
          {
            "update": 477,
            "environment_steps": 1953792,
            "curriculum": 3,
            "completed_episodes": 3513,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.3149438798427582,
            "loss": 6.741808250546455,
            "approx_kl": 0.006072277214116184,
            "wall_seconds": 2402.0557704000094
          },
          {
            "update": 478,
            "environment_steps": 1957888,
            "curriculum": 3,
            "completed_episodes": 3520,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.2870480716228485,
            "loss": 10.751126855611801,
            "approx_kl": 0.002791908918879926,
            "wall_seconds": 2409.9712368000182
          },
          {
            "update": 479,
            "environment_steps": 1961984,
            "curriculum": 3,
            "completed_episodes": 3525,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.3243725299835205,
            "loss": 5.637631803750992,
            "approx_kl": 0.0037141096872801427,
            "wall_seconds": 2417.7626237000222
          },
          {
            "update": 480,
            "environment_steps": 1966080,
            "curriculum": 3,
            "completed_episodes": 3529,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.28051045536994934,
            "loss": 4.395832002162933,
            "approx_kl": 0.004900386156805325,
            "wall_seconds": 2425.679220400023
          },
          {
            "update": 481,
            "environment_steps": 1970176,
            "curriculum": 3,
            "completed_episodes": 3533,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.27667030692100525,
            "loss": 6.3666664361953735,
            "approx_kl": 0.003867423645715462,
            "wall_seconds": 2433.7203282000264
          },
          {
            "update": 482,
            "environment_steps": 1974272,
            "curriculum": 3,
            "completed_episodes": 3538,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2759557068347931,
            "loss": 7.001233547925949,
            "approx_kl": 0.005438325002614874,
            "wall_seconds": 2441.671435600001
          },
          {
            "update": 483,
            "environment_steps": 1978368,
            "curriculum": 3,
            "completed_episodes": 3542,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2495408058166504,
            "loss": 5.322318375110626,
            "approx_kl": 0.0036245104820409324,
            "wall_seconds": 2449.7797693000175
          },
          {
            "update": 484,
            "environment_steps": 1982464,
            "curriculum": 3,
            "completed_episodes": 3544,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3212834596633911,
            "loss": 2.0798442512750626,
            "approx_kl": 0.005168156756553799,
            "wall_seconds": 2457.781209800014
          },
          {
            "update": 485,
            "environment_steps": 1986560,
            "curriculum": 3,
            "completed_episodes": 3548,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3178222179412842,
            "loss": 5.215975403785706,
            "approx_kl": 0.0026215450598101597,
            "wall_seconds": 2465.7546532000124
          },
          {
            "update": 486,
            "environment_steps": 1990656,
            "curriculum": 3,
            "completed_episodes": 3551,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29136964678764343,
            "loss": 4.532879024744034,
            "approx_kl": 0.0033083458874898497,
            "wall_seconds": 2473.843424600025
          },
          {
            "update": 487,
            "environment_steps": 1994752,
            "curriculum": 3,
            "completed_episodes": 3556,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.27632737159729004,
            "loss": 4.475130632519722,
            "approx_kl": 0.005744742429669714,
            "wall_seconds": 2481.8182706000225
          },
          {
            "update": 488,
            "environment_steps": 1998848,
            "curriculum": 3,
            "completed_episodes": 3562,
            "new_episodes": 6,
            "new_successes": 3,
            "mean_step_reward": -0.2534797191619873,
            "loss": 5.414409190416336,
            "approx_kl": 0.003062568481254857,
            "wall_seconds": 2489.9723708000092
          },
          {
            "update": 489,
            "environment_steps": 2002944,
            "curriculum": 3,
            "completed_episodes": 3568,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2744378447532654,
            "loss": 5.427227601408958,
            "approx_kl": 0.003085297725192504,
            "wall_seconds": 2497.9075176000188
          },
          {
            "update": 490,
            "environment_steps": 2007040,
            "curriculum": 3,
            "completed_episodes": 3574,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.2964242100715637,
            "loss": 5.8142701387405396,
            "approx_kl": 0.004811062968656188,
            "wall_seconds": 2505.9827981000126
          },
          {
            "update": 491,
            "environment_steps": 2011136,
            "curriculum": 3,
            "completed_episodes": 3577,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2728514075279236,
            "loss": 3.552677199244499,
            "approx_kl": 0.0066656664821493905,
            "wall_seconds": 2513.989037500025
          },
          {
            "update": 492,
            "environment_steps": 2015232,
            "curriculum": 3,
            "completed_episodes": 3580,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29697853326797485,
            "loss": 3.6572786271572113,
            "approx_kl": 0.005938728507317137,
            "wall_seconds": 2522.115249900002
          },
          {
            "update": 493,
            "environment_steps": 2019328,
            "curriculum": 3,
            "completed_episodes": 3584,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3140096962451935,
            "loss": 4.323298826813698,
            "approx_kl": 0.004176667214778718,
            "wall_seconds": 2530.0785978000204
          },
          {
            "update": 494,
            "environment_steps": 2023424,
            "curriculum": 3,
            "completed_episodes": 3586,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3150053322315216,
            "loss": 2.1951526775956154,
            "approx_kl": 0.007930653962830547,
            "wall_seconds": 2538.0782238000247
          },
          {
            "update": 495,
            "environment_steps": 2027520,
            "curriculum": 3,
            "completed_episodes": 3589,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2787788510322571,
            "loss": 3.3466123044490814,
            "approx_kl": 0.00424984884011792,
            "wall_seconds": 2546.1326136999996
          },
          {
            "update": 496,
            "environment_steps": 2031616,
            "curriculum": 3,
            "completed_episodes": 3591,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.30394846200942993,
            "loss": 3.4001133888959885,
            "approx_kl": 0.006849048037111061,
            "wall_seconds": 2554.048701700027
          },
          {
            "update": 497,
            "environment_steps": 2035712,
            "curriculum": 3,
            "completed_episodes": 3598,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.29566627740859985,
            "loss": 9.591462522745132,
            "approx_kl": 0.002941234448371688,
            "wall_seconds": 2562.1409104000195
          },
          {
            "update": 498,
            "environment_steps": 2039808,
            "curriculum": 3,
            "completed_episodes": 3604,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.279325932264328,
            "loss": 6.4530819952487946,
            "approx_kl": 0.004002852725534467,
            "wall_seconds": 2570.2062557000027
          },
          {
            "update": 499,
            "environment_steps": 2043904,
            "curriculum": 3,
            "completed_episodes": 3607,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3000320792198181,
            "loss": 3.210026577115059,
            "approx_kl": 0.005995652620185865,
            "wall_seconds": 2578.2189679000003
          },
          {
            "update": 500,
            "environment_steps": 2048000,
            "curriculum": 3,
            "completed_episodes": 3612,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2986007034778595,
            "loss": 5.810072660446167,
            "approx_kl": 0.003316031459689839,
            "wall_seconds": 2586.251694800012
          },
          {
            "update": 501,
            "environment_steps": 2052096,
            "curriculum": 3,
            "completed_episodes": 3616,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.24579697847366333,
            "loss": 3.1612935215234756,
            "approx_kl": 0.005014789283450227,
            "wall_seconds": 2593.3839832000085
          },
          {
            "update": 502,
            "environment_steps": 2056192,
            "curriculum": 3,
            "completed_episodes": 3621,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2997571527957916,
            "loss": 5.1127873957157135,
            "approx_kl": 0.005132045043865219,
            "wall_seconds": 2599.0101793000067
          },
          {
            "update": 503,
            "environment_steps": 2060288,
            "curriculum": 3,
            "completed_episodes": 3622,
            "new_episodes": 1,
            "new_successes": 1,
            "mean_step_reward": -0.26027658581733704,
            "loss": 4.213446073234081,
            "approx_kl": 0.006255686159420293,
            "wall_seconds": 2606.7998627000197
          },
          {
            "update": 504,
            "environment_steps": 2064384,
            "curriculum": 3,
            "completed_episodes": 3625,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.30219006538391113,
            "loss": 2.6369296684861183,
            "approx_kl": 0.0027725471518351696,
            "wall_seconds": 2615.9213728000177
          },
          {
            "update": 505,
            "environment_steps": 2068480,
            "curriculum": 3,
            "completed_episodes": 3629,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3182867765426636,
            "loss": 4.819591820240021,
            "approx_kl": 0.003469948860583827,
            "wall_seconds": 2622.8265725000238
          },
          {
            "update": 506,
            "environment_steps": 2072576,
            "curriculum": 3,
            "completed_episodes": 3633,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3053073287010193,
            "loss": 4.94990773499012,
            "approx_kl": 0.007095947989000706,
            "wall_seconds": 2631.8267164000135
          },
          {
            "update": 507,
            "environment_steps": 2076672,
            "curriculum": 3,
            "completed_episodes": 3640,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2630574703216553,
            "loss": 6.66786926984787,
            "approx_kl": 0.004563653594232164,
            "wall_seconds": 2640.9332855000102
          },
          {
            "update": 508,
            "environment_steps": 2080768,
            "curriculum": 3,
            "completed_episodes": 3643,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.268701434135437,
            "loss": 3.8741452395915985,
            "approx_kl": 0.008351761625817744,
            "wall_seconds": 2650.097687200003
          },
          {
            "update": 509,
            "environment_steps": 2084864,
            "curriculum": 3,
            "completed_episodes": 3647,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.30534613132476807,
            "loss": 4.894015654921532,
            "approx_kl": 0.005074575536127668,
            "wall_seconds": 2659.258493900008
          },
          {
            "update": 510,
            "environment_steps": 2088960,
            "curriculum": 3,
            "completed_episodes": 3650,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.269975483417511,
            "loss": 4.344250828027725,
            "approx_kl": 0.0026645217803888954,
            "wall_seconds": 2668.4441065000137
          },
          {
            "update": 511,
            "environment_steps": 2093056,
            "curriculum": 3,
            "completed_episodes": 3654,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2577959895133972,
            "loss": 5.425987869501114,
            "approx_kl": 0.0040749377803877,
            "wall_seconds": 2677.735221800016
          },
          {
            "update": 512,
            "environment_steps": 2097152,
            "curriculum": 3,
            "completed_episodes": 3658,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2679177224636078,
            "loss": 4.494501769542694,
            "approx_kl": 0.003879039330058731,
            "wall_seconds": 2686.9585826000257
          },
          {
            "update": 513,
            "environment_steps": 2101248,
            "curriculum": 3,
            "completed_episodes": 3660,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3158718943595886,
            "loss": 3.033582352101803,
            "approx_kl": 0.0042364355285826605,
            "wall_seconds": 2694.320962800004
          },
          {
            "update": 514,
            "environment_steps": 2105344,
            "curriculum": 3,
            "completed_episodes": 3665,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.3091224730014801,
            "loss": 5.4309555143117905,
            "approx_kl": 0.007485266265575774,
            "wall_seconds": 2702.8557896000275
          },
          {
            "update": 515,
            "environment_steps": 2109440,
            "curriculum": 3,
            "completed_episodes": 3667,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.30835914611816406,
            "loss": 2.527988538146019,
            "approx_kl": 0.004425608422025107,
            "wall_seconds": 2711.941781900008
          },
          {
            "update": 516,
            "environment_steps": 2113536,
            "curriculum": 3,
            "completed_episodes": 3673,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.270987331867218,
            "loss": 5.995154708623886,
            "approx_kl": 0.0048700310217100196,
            "wall_seconds": 2721.1196579000098
          },
          {
            "update": 517,
            "environment_steps": 2117632,
            "curriculum": 3,
            "completed_episodes": 3678,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.28296294808387756,
            "loss": 6.594939082860947,
            "approx_kl": 0.00811369327493594,
            "wall_seconds": 2730.294518800016
          },
          {
            "update": 518,
            "environment_steps": 2121728,
            "curriculum": 3,
            "completed_episodes": 3681,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2871611714363098,
            "loss": 3.6748079359531403,
            "approx_kl": 0.008029793170862831,
            "wall_seconds": 2739.496194300009
          },
          {
            "update": 519,
            "environment_steps": 2125824,
            "curriculum": 3,
            "completed_episodes": 3686,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2782086133956909,
            "loss": 4.946107059717178,
            "approx_kl": 0.007522047308157198,
            "wall_seconds": 2748.7310201000073
          },
          {
            "update": 520,
            "environment_steps": 2129920,
            "curriculum": 3,
            "completed_episodes": 3690,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.32535383105278015,
            "loss": 4.165965050458908,
            "approx_kl": 0.008157093689078465,
            "wall_seconds": 2757.881676000019
          },
          {
            "update": 521,
            "environment_steps": 2134016,
            "curriculum": 3,
            "completed_episodes": 3694,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.32536619901657104,
            "loss": 4.13493449985981,
            "approx_kl": 0.010248164438962704,
            "wall_seconds": 2766.022620500007
          },
          {
            "update": 522,
            "environment_steps": 2138112,
            "curriculum": 3,
            "completed_episodes": 3695,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.29055696725845337,
            "loss": 1.4018033295869827,
            "approx_kl": 0.008284075312985806,
            "wall_seconds": 2772.2607386000163
          },
          {
            "update": 523,
            "environment_steps": 2142208,
            "curriculum": 3,
            "completed_episodes": 3698,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.29638946056365967,
            "loss": 3.41007748991251,
            "approx_kl": 0.006641207462962484,
            "wall_seconds": 2780.0652889000194
          },
          {
            "update": 524,
            "environment_steps": 2146304,
            "curriculum": 3,
            "completed_episodes": 3701,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3219902217388153,
            "loss": 3.8646604120731354,
            "approx_kl": 0.006496213984064525,
            "wall_seconds": 2787.9057603000256
          },
          {
            "update": 525,
            "environment_steps": 2150400,
            "curriculum": 3,
            "completed_episodes": 3706,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2872081398963928,
            "loss": 7.268525242805481,
            "approx_kl": 0.005613122782961,
            "wall_seconds": 2795.7298184000247
          },
          {
            "update": 526,
            "environment_steps": 2154496,
            "curriculum": 3,
            "completed_episodes": 3711,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.25031930208206177,
            "loss": 7.244561433792114,
            "approx_kl": 0.004173088756942889,
            "wall_seconds": 2803.626180000021
          },
          {
            "update": 527,
            "environment_steps": 2158592,
            "curriculum": 3,
            "completed_episodes": 3716,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.24547071754932404,
            "loss": 8.883065432310104,
            "approx_kl": 0.0025446071522310376,
            "wall_seconds": 2811.3907110000146
          },
          {
            "update": 528,
            "environment_steps": 2162688,
            "curriculum": 3,
            "completed_episodes": 3720,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.25744616985321045,
            "loss": 4.475535735487938,
            "approx_kl": 0.0035866536054527387,
            "wall_seconds": 2819.07073040001
          },
          {
            "update": 529,
            "environment_steps": 2166784,
            "curriculum": 3,
            "completed_episodes": 3723,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.29465997219085693,
            "loss": 3.43299400806427,
            "approx_kl": 0.004876640825386858,
            "wall_seconds": 2826.8451432000147
          },
          {
            "update": 530,
            "environment_steps": 2170880,
            "curriculum": 3,
            "completed_episodes": 3725,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2594062089920044,
            "loss": 3.6432855501770973,
            "approx_kl": 0.006274966395722004,
            "wall_seconds": 2834.646883700014
          },
          {
            "update": 531,
            "environment_steps": 2174976,
            "curriculum": 3,
            "completed_episodes": 3729,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.27680888772010803,
            "loss": 4.861864387989044,
            "approx_kl": 0.0033777907556213904,
            "wall_seconds": 2842.4293976000044
          },
          {
            "update": 532,
            "environment_steps": 2179072,
            "curriculum": 3,
            "completed_episodes": 3732,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3112165331840515,
            "loss": 2.873491734266281,
            "approx_kl": 0.006773188972147182,
            "wall_seconds": 2850.329857900011
          },
          {
            "update": 533,
            "environment_steps": 2183168,
            "curriculum": 3,
            "completed_episodes": 3736,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.28163543343544006,
            "loss": 4.581849038600922,
            "approx_kl": 0.011442915911175078,
            "wall_seconds": 2858.3291696000088
          },
          {
            "update": 534,
            "environment_steps": 2187264,
            "curriculum": 3,
            "completed_episodes": 3739,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.30584216117858887,
            "loss": 3.809432625770569,
            "approx_kl": 0.004858367814449593,
            "wall_seconds": 2866.310581800004
          },
          {
            "update": 535,
            "environment_steps": 2191360,
            "curriculum": 3,
            "completed_episodes": 3745,
            "new_episodes": 6,
            "new_successes": 1,
            "mean_step_reward": -0.28618985414505005,
            "loss": 5.49289083480835,
            "approx_kl": 0.0069594178421539254,
            "wall_seconds": 2874.3271072000207
          },
          {
            "update": 536,
            "environment_steps": 2195456,
            "curriculum": 3,
            "completed_episodes": 3750,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.2613838315010071,
            "loss": 5.340615898370743,
            "approx_kl": 0.0064195059312623926,
            "wall_seconds": 2882.447880200023
          },
          {
            "update": 537,
            "environment_steps": 2199552,
            "curriculum": 3,
            "completed_episodes": 3756,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.23522737622261047,
            "loss": 7.380579888820648,
            "approx_kl": 0.003066428060265025,
            "wall_seconds": 2890.5988822000218
          },
          {
            "update": 538,
            "environment_steps": 2203648,
            "curriculum": 3,
            "completed_episodes": 3761,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.26567548513412476,
            "loss": 5.7109991908073425,
            "approx_kl": 0.002677254095033277,
            "wall_seconds": 2898.622218900011
          },
          {
            "update": 539,
            "environment_steps": 2207744,
            "curriculum": 3,
            "completed_episodes": 3767,
            "new_episodes": 6,
            "new_successes": 3,
            "mean_step_reward": -0.29374533891677856,
            "loss": 6.1751898676157,
            "approx_kl": 0.004621658557880437,
            "wall_seconds": 2906.7227690000145
          },
          {
            "update": 540,
            "environment_steps": 2211840,
            "curriculum": 3,
            "completed_episodes": 3768,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.3051919937133789,
            "loss": 1.6515031605958939,
            "approx_kl": 0.0058829036315728445,
            "wall_seconds": 2914.8270157000225
          },
          {
            "update": 541,
            "environment_steps": 2215936,
            "curriculum": 3,
            "completed_episodes": 3770,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.2898717522621155,
            "loss": 2.6832712292671204,
            "approx_kl": 0.005377857811254216,
            "wall_seconds": 2922.9498218000226
          },
          {
            "update": 542,
            "environment_steps": 2220032,
            "curriculum": 3,
            "completed_episodes": 3775,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.25741317868232727,
            "loss": 5.220525607466698,
            "approx_kl": 0.006615647769649513,
            "wall_seconds": 2931.1174037000164
          },
          {
            "update": 543,
            "environment_steps": 2224128,
            "curriculum": 3,
            "completed_episodes": 3775,
            "new_episodes": 0,
            "new_successes": 0,
            "mean_step_reward": -0.264986127614975,
            "loss": 1.4782568365335464,
            "approx_kl": 0.008170171269739512,
            "wall_seconds": 2939.310726100026
          },
          {
            "update": 544,
            "environment_steps": 2228224,
            "curriculum": 3,
            "completed_episodes": 3781,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.3217356204986572,
            "loss": 6.760446935892105,
            "approx_kl": 0.004847020572924521,
            "wall_seconds": 2947.403531900025
          },
          {
            "update": 545,
            "environment_steps": 2232320,
            "curriculum": 3,
            "completed_episodes": 3784,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2884739637374878,
            "loss": 3.8414334654808044,
            "approx_kl": 0.006183849534863839,
            "wall_seconds": 2955.508842900017
          },
          {
            "update": 546,
            "environment_steps": 2236416,
            "curriculum": 3,
            "completed_episodes": 3790,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.29119884967803955,
            "loss": 5.377977699041367,
            "approx_kl": 0.003357667796080932,
            "wall_seconds": 2963.567910300015
          },
          {
            "update": 547,
            "environment_steps": 2240512,
            "curriculum": 3,
            "completed_episodes": 3794,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2756264805793762,
            "loss": 4.617246776819229,
            "approx_kl": 0.004189293569652364,
            "wall_seconds": 2971.6844562000188
          },
          {
            "update": 548,
            "environment_steps": 2244608,
            "curriculum": 3,
            "completed_episodes": 3801,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2910540699958801,
            "loss": 5.401969939470291,
            "approx_kl": 0.003229712543543428,
            "wall_seconds": 2979.8388879000267
          },
          {
            "update": 549,
            "environment_steps": 2248704,
            "curriculum": 3,
            "completed_episodes": 3804,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.23427079617977142,
            "loss": 4.087374970316887,
            "approx_kl": 0.006554099705681438,
            "wall_seconds": 2987.9613467000017
          },
          {
            "update": 550,
            "environment_steps": 2252800,
            "curriculum": 3,
            "completed_episodes": 3807,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.26683691143989563,
            "loss": 3.6579644978046417,
            "approx_kl": 0.005224602482485352,
            "wall_seconds": 2996.1256647000264
          },
          {
            "update": 551,
            "environment_steps": 2256896,
            "curriculum": 3,
            "completed_episodes": 3810,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.31432783603668213,
            "loss": 3.9079203605651855,
            "approx_kl": 0.004470775566005614,
            "wall_seconds": 3004.3029100999993
          },
          {
            "update": 552,
            "environment_steps": 2260992,
            "curriculum": 3,
            "completed_episodes": 3813,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.30333632230758667,
            "loss": 3.8064556568861008,
            "approx_kl": 0.0070890989336476196,
            "wall_seconds": 3012.4105400000117
          },
          {
            "update": 553,
            "environment_steps": 2265088,
            "curriculum": 3,
            "completed_episodes": 3816,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.309101939201355,
            "loss": 4.286187782883644,
            "approx_kl": 0.0102483617629332,
            "wall_seconds": 3020.37708950002
          },
          {
            "update": 554,
            "environment_steps": 2269184,
            "curriculum": 3,
            "completed_episodes": 3820,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.30874723196029663,
            "loss": 4.8691665828228,
            "approx_kl": 0.014001205538079375,
            "wall_seconds": 3028.515574800025
          },
          {
            "update": 555,
            "environment_steps": 2273280,
            "curriculum": 3,
            "completed_episodes": 3828,
            "new_episodes": 8,
            "new_successes": 2,
            "mean_step_reward": -0.2659713923931122,
            "loss": 9.608015090227127,
            "approx_kl": 0.004882115106738638,
            "wall_seconds": 3035.45747340002
          },
          {
            "update": 556,
            "environment_steps": 2277376,
            "curriculum": 3,
            "completed_episodes": 3830,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.23367170989513397,
            "loss": 3.3149490505456924,
            "approx_kl": 0.0049162173199874815,
            "wall_seconds": 3041.173884900025
          },
          {
            "update": 557,
            "environment_steps": 2281472,
            "curriculum": 3,
            "completed_episodes": 3834,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2537708282470703,
            "loss": 4.625955700874329,
            "approx_kl": 0.0034654617265914567,
            "wall_seconds": 3049.0803841000015
          },
          {
            "update": 558,
            "environment_steps": 2285568,
            "curriculum": 3,
            "completed_episodes": 3840,
            "new_episodes": 6,
            "new_successes": 2,
            "mean_step_reward": -0.2722655236721039,
            "loss": 5.2293287217617035,
            "approx_kl": 0.006092020918003982,
            "wall_seconds": 3058.1543378000206
          },
          {
            "update": 559,
            "environment_steps": 2289664,
            "curriculum": 3,
            "completed_episodes": 3843,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.305589497089386,
            "loss": 3.047100216150284,
            "approx_kl": 0.006900689924805192,
            "wall_seconds": 3064.9728825000057
          },
          {
            "update": 560,
            "environment_steps": 2293760,
            "curriculum": 3,
            "completed_episodes": 3846,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.28159505128860474,
            "loss": 3.2426136434078217,
            "approx_kl": 0.007947346170112723,
            "wall_seconds": 3073.9795735000225
          },
          {
            "update": 561,
            "environment_steps": 2297856,
            "curriculum": 3,
            "completed_episodes": 3851,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.28654974699020386,
            "loss": 5.077508166432381,
            "approx_kl": 0.00604751578066498,
            "wall_seconds": 3083.0840994000027
          },
          {
            "update": 562,
            "environment_steps": 2301952,
            "curriculum": 3,
            "completed_episodes": 3855,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.30747824907302856,
            "loss": 4.837609380483627,
            "approx_kl": 0.0026130144506169017,
            "wall_seconds": 3092.1574538999994
          },
          {
            "update": 563,
            "environment_steps": 2306048,
            "curriculum": 3,
            "completed_episodes": 3859,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2678561806678772,
            "loss": 4.706837043166161,
            "approx_kl": 0.004928701422613813,
            "wall_seconds": 3101.3618836000096
          },
          {
            "update": 564,
            "environment_steps": 2310144,
            "curriculum": 3,
            "completed_episodes": 3861,
            "new_episodes": 2,
            "new_successes": 2,
            "mean_step_reward": -0.24733871221542358,
            "loss": 1.4921094998717308,
            "approx_kl": 0.006253886316699209,
            "wall_seconds": 3110.588243300008
          },
          {
            "update": 565,
            "environment_steps": 2314240,
            "curriculum": 3,
            "completed_episodes": 3870,
            "new_episodes": 9,
            "new_successes": 2,
            "mean_step_reward": -0.28955814242362976,
            "loss": 11.25325059890747,
            "approx_kl": 0.004157996281719534,
            "wall_seconds": 3119.87111550002
          },
          {
            "update": 566,
            "environment_steps": 2318336,
            "curriculum": 3,
            "completed_episodes": 3873,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3201613426208496,
            "loss": 4.089864134788513,
            "approx_kl": 0.0026323199235775974,
            "wall_seconds": 3129.089533600025
          },
          {
            "update": 567,
            "environment_steps": 2322432,
            "curriculum": 3,
            "completed_episodes": 3877,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.28037363290786743,
            "loss": 4.525062024593353,
            "approx_kl": 0.005597606184892356,
            "wall_seconds": 3136.5510342000052
          },
          {
            "update": 568,
            "environment_steps": 2326528,
            "curriculum": 3,
            "completed_episodes": 3880,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.2598862648010254,
            "loss": 4.018755853176117,
            "approx_kl": 0.005310080312483478,
            "wall_seconds": 3144.9758784000005
          },
          {
            "update": 569,
            "environment_steps": 2330624,
            "curriculum": 3,
            "completed_episodes": 3884,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2983536124229431,
            "loss": 3.0996828004717827,
            "approx_kl": 0.005444073805847438,
            "wall_seconds": 3153.957143800013
          },
          {
            "update": 570,
            "environment_steps": 2334720,
            "curriculum": 3,
            "completed_episodes": 3891,
            "new_episodes": 7,
            "new_successes": 2,
            "mean_step_reward": -0.30241313576698303,
            "loss": 8.478412866592407,
            "approx_kl": 0.0017973588219319936,
            "wall_seconds": 3163.0936442000093
          },
          {
            "update": 571,
            "environment_steps": 2338816,
            "curriculum": 3,
            "completed_episodes": 3892,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.31960439682006836,
            "loss": 1.6983257159590721,
            "approx_kl": 0.004120111480006017,
            "wall_seconds": 3172.2987242000236
          },
          {
            "update": 572,
            "environment_steps": 2342912,
            "curriculum": 3,
            "completed_episodes": 3896,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3195854127407074,
            "loss": 3.6480655521154404,
            "approx_kl": 0.005065224162535742,
            "wall_seconds": 3181.461837700015
          },
          {
            "update": 573,
            "environment_steps": 2347008,
            "curriculum": 3,
            "completed_episodes": 3898,
            "new_episodes": 2,
            "new_successes": 0,
            "mean_step_reward": -0.3052048683166504,
            "loss": 2.860568881034851,
            "approx_kl": 0.006804564018239034,
            "wall_seconds": 3190.6492986000085
          },
          {
            "update": 574,
            "environment_steps": 2351104,
            "curriculum": 3,
            "completed_episodes": 3905,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.27464181184768677,
            "loss": 7.70151561498642,
            "approx_kl": 0.0027202286510146223,
            "wall_seconds": 3199.933402900002
          },
          {
            "update": 575,
            "environment_steps": 2355200,
            "curriculum": 3,
            "completed_episodes": 3910,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2608758807182312,
            "loss": 7.262127935886383,
            "approx_kl": 0.0032006699257181026,
            "wall_seconds": 3208.1186748000036
          },
          {
            "update": 576,
            "environment_steps": 2359296,
            "curriculum": 3,
            "completed_episodes": 3912,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.32060348987579346,
            "loss": 3.3241966515779495,
            "approx_kl": 0.004556625375698786,
            "wall_seconds": 3213.717213900003
          },
          {
            "update": 577,
            "environment_steps": 2363392,
            "curriculum": 3,
            "completed_episodes": 3915,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.31718599796295166,
            "loss": 3.4329576045274734,
            "approx_kl": 0.0018500215592212044,
            "wall_seconds": 3219.369262500026
          },
          {
            "update": 578,
            "environment_steps": 2367488,
            "curriculum": 3,
            "completed_episodes": 3919,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.2962181568145752,
            "loss": 4.541822135448456,
            "approx_kl": 0.008181110606528819,
            "wall_seconds": 3225.052245900006
          },
          {
            "update": 579,
            "environment_steps": 2371584,
            "curriculum": 3,
            "completed_episodes": 3924,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.31713640689849854,
            "loss": 5.325452923774719,
            "approx_kl": 0.004039074392494513,
            "wall_seconds": 3230.749331500003
          },
          {
            "update": 580,
            "environment_steps": 2375680,
            "curriculum": 3,
            "completed_episodes": 3927,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3096867501735687,
            "loss": 4.231264263391495,
            "approx_kl": 0.003738537085155258,
            "wall_seconds": 3236.4263309000235
          },
          {
            "update": 581,
            "environment_steps": 2379776,
            "curriculum": 3,
            "completed_episodes": 3931,
            "new_episodes": 4,
            "new_successes": 2,
            "mean_step_reward": -0.2765393853187561,
            "loss": 4.29595547914505,
            "approx_kl": 0.005950167265837081,
            "wall_seconds": 3242.187363600009
          },
          {
            "update": 582,
            "environment_steps": 2383872,
            "curriculum": 3,
            "completed_episodes": 3934,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.29864564538002014,
            "loss": 3.116626091301441,
            "approx_kl": 0.003582393459510058,
            "wall_seconds": 3247.965281600016
          },
          {
            "update": 583,
            "environment_steps": 2387968,
            "curriculum": 3,
            "completed_episodes": 3938,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3030596375465393,
            "loss": 4.482811436057091,
            "approx_kl": 0.006007509618939366,
            "wall_seconds": 3253.689725700009
          },
          {
            "update": 584,
            "environment_steps": 2392064,
            "curriculum": 3,
            "completed_episodes": 3945,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.28619951009750366,
            "loss": 9.474295914173126,
            "approx_kl": 0.0044578653541975655,
            "wall_seconds": 3259.452731700003
          },
          {
            "update": 585,
            "environment_steps": 2396160,
            "curriculum": 3,
            "completed_episodes": 3949,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.2593385577201843,
            "loss": 3.768128052353859,
            "approx_kl": 0.0036326384361018427,
            "wall_seconds": 3265.1972275000007
          },
          {
            "update": 586,
            "environment_steps": 2400256,
            "curriculum": 3,
            "completed_episodes": 3954,
            "new_episodes": 5,
            "new_successes": 2,
            "mean_step_reward": -0.2613697350025177,
            "loss": 4.127467915415764,
            "approx_kl": 0.005720597298932262,
            "wall_seconds": 3270.970676500001
          },
          {
            "update": 587,
            "environment_steps": 2404352,
            "curriculum": 3,
            "completed_episodes": 3958,
            "new_episodes": 4,
            "new_successes": 1,
            "mean_step_reward": -0.3006589412689209,
            "loss": 4.200466200709343,
            "approx_kl": 0.005732590056140907,
            "wall_seconds": 3276.6428895000136
          },
          {
            "update": 588,
            "environment_steps": 2408448,
            "curriculum": 3,
            "completed_episodes": 3961,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2731321454048157,
            "loss": 4.395112156867981,
            "approx_kl": 0.00859777396908612,
            "wall_seconds": 3282.396998100012
          },
          {
            "update": 589,
            "environment_steps": 2412544,
            "curriculum": 3,
            "completed_episodes": 3967,
            "new_episodes": 6,
            "new_successes": 0,
            "mean_step_reward": -0.32647544145584106,
            "loss": 6.572574943304062,
            "approx_kl": 0.004195978450297844,
            "wall_seconds": 3288.0948659999995
          },
          {
            "update": 590,
            "environment_steps": 2416640,
            "curriculum": 3,
            "completed_episodes": 3969,
            "new_episodes": 2,
            "new_successes": 1,
            "mean_step_reward": -0.30588600039482117,
            "loss": 3.6165490597486496,
            "approx_kl": 0.004751867360027973,
            "wall_seconds": 3293.8332954000216
          },
          {
            "update": 591,
            "environment_steps": 2420736,
            "curriculum": 3,
            "completed_episodes": 3972,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3205541670322418,
            "loss": 3.5467133671045303,
            "approx_kl": 0.00394800537469564,
            "wall_seconds": 3299.612131300004
          },
          {
            "update": 592,
            "environment_steps": 2424832,
            "curriculum": 3,
            "completed_episodes": 3975,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.31408700346946716,
            "loss": 2.3276767432689667,
            "approx_kl": 0.004068996480782516,
            "wall_seconds": 3305.3069828000152
          },
          {
            "update": 593,
            "environment_steps": 2428928,
            "curriculum": 3,
            "completed_episodes": 3980,
            "new_episodes": 5,
            "new_successes": 0,
            "mean_step_reward": -0.3149118423461914,
            "loss": 6.087741732597351,
            "approx_kl": 0.004405289815622382,
            "wall_seconds": 3310.9727905000036
          },
          {
            "update": 594,
            "environment_steps": 2433024,
            "curriculum": 3,
            "completed_episodes": 3985,
            "new_episodes": 5,
            "new_successes": 1,
            "mean_step_reward": -0.2448635697364807,
            "loss": 5.91727089881897,
            "approx_kl": 0.002893942564696772,
            "wall_seconds": 3316.723160500027
          },
          {
            "update": 595,
            "environment_steps": 2437120,
            "curriculum": 3,
            "completed_episodes": 3988,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3104824423789978,
            "loss": 3.9167684763669968,
            "approx_kl": 0.003465737750957487,
            "wall_seconds": 3322.4587290000054
          },
          {
            "update": 596,
            "environment_steps": 2441216,
            "curriculum": 3,
            "completed_episodes": 3992,
            "new_episodes": 4,
            "new_successes": 0,
            "mean_step_reward": -0.3221820890903473,
            "loss": 5.2107003480196,
            "approx_kl": 0.004236448268784443,
            "wall_seconds": 3328.1461860000272
          },
          {
            "update": 597,
            "environment_steps": 2445312,
            "curriculum": 3,
            "completed_episodes": 3995,
            "new_episodes": 3,
            "new_successes": 1,
            "mean_step_reward": -0.2965158522129059,
            "loss": 5.066319316625595,
            "approx_kl": 0.006258475594222546,
            "wall_seconds": 3333.8816551000054
          },
          {
            "update": 598,
            "environment_steps": 2449408,
            "curriculum": 3,
            "completed_episodes": 4002,
            "new_episodes": 7,
            "new_successes": 1,
            "mean_step_reward": -0.2784538269042969,
            "loss": 7.210305869579315,
            "approx_kl": 0.003308263621875085,
            "wall_seconds": 3339.6813486000174
          },
          {
            "update": 599,
            "environment_steps": 2453504,
            "curriculum": 3,
            "completed_episodes": 4003,
            "new_episodes": 1,
            "new_successes": 0,
            "mean_step_reward": -0.28750571608543396,
            "loss": 2.0612469874322414,
            "approx_kl": 0.002918079942901386,
            "wall_seconds": 3345.3841806000273
          },
          {
            "update": 600,
            "environment_steps": 2457600,
            "curriculum": 3,
            "completed_episodes": 4006,
            "new_episodes": 3,
            "new_successes": 0,
            "mean_step_reward": -0.3065827190876007,
            "loss": 4.510078027844429,
            "approx_kl": 0.0054704084832337685,
            "wall_seconds": 3351.1289456000086
          }
        ],
        "completed_episodes": 4006
      },
      "outputs/online/training-v6-broad-w1024.json": {
        "sha256": "800a11e7f0c4db44e3922380f02254f45afa87e9735c34f4b3e8b47abe4c1fdb",
        "metadata": {
          "method": "Supervised teacher examples; DAgger when aggregated datasets are supplied",
          "inputs": "Fresh depth image, continuous current XYZ and yaw, continuous goal XYZ",
          "parameters": 1793538,
          "width": 1024,
          "training_examples": 950000,
          "validation_examples": 50000,
          "steps": 24000,
          "batch": 2048,
          "seed": 130941,
          "data": [
            "outputs/online/imitation-initial.npz",
            "outputs/online/imitation-broad.npz"
          ],
          "resume": null,
          "learning_rate": 0.001,
          "inference": "Fresh network forward each 100 ms; no map-output cache, no route planner, no NAV IDs in policy",
          "validation_scope": "Image split is a fitting diagnostic only; navigation validation uses separate development situations."
        },
        "trace": [
          {
            "step": 1000,
            "loss": 0.13455131649971008,
            "image_validation_mae_degrees": 24.260234832763672,
            "image_validation_p95_degrees": 72.40816497802734,
            "wall_seconds": 2.6864100999955554
          },
          {
            "step": 2000,
            "loss": 0.10482844710350037,
            "image_validation_mae_degrees": 15.389294624328613,
            "image_validation_p95_degrees": 44.21363067626953,
            "wall_seconds": 4.845928800001275
          },
          {
            "step": 3000,
            "loss": 0.08564242720603943,
            "image_validation_mae_degrees": 12.805384635925293,
            "image_validation_p95_degrees": 37.08934783935547,
            "wall_seconds": 6.894914000004064
          },
          {
            "step": 4000,
            "loss": 0.0542885959148407,
            "image_validation_mae_degrees": 11.5230131149292,
            "image_validation_p95_degrees": 32.79106903076172,
            "wall_seconds": 9.012456399999792
          },
          {
            "step": 5000,
            "loss": 0.04559452086687088,
            "image_validation_mae_degrees": 11.137199401855469,
            "image_validation_p95_degrees": 31.91551399230957,
            "wall_seconds": 11.142477899993537
          },
          {
            "step": 6000,
            "loss": 0.03484624996781349,
            "image_validation_mae_degrees": 10.454090118408203,
            "image_validation_p95_degrees": 29.913537979125977,
            "wall_seconds": 13.197855700011132
          },
          {
            "step": 7000,
            "loss": 0.025967027992010117,
            "image_validation_mae_degrees": 10.04281997680664,
            "image_validation_p95_degrees": 29.036226272583008,
            "wall_seconds": 15.31066039999132
          },
          {
            "step": 8000,
            "loss": 0.020914237946271896,
            "image_validation_mae_degrees": 9.722368240356445,
            "image_validation_p95_degrees": 28.15461540222168,
            "wall_seconds": 17.30012950001401
          },
          {
            "step": 9000,
            "loss": 0.016769705340266228,
            "image_validation_mae_degrees": 9.522614479064941,
            "image_validation_p95_degrees": 27.597930908203125,
            "wall_seconds": 19.47322179999901
          },
          {
            "step": 10000,
            "loss": 0.013860784471035004,
            "image_validation_mae_degrees": 9.291153907775879,
            "image_validation_p95_degrees": 27.489791870117188,
            "wall_seconds": 21.494420200004242
          },
          {
            "step": 11000,
            "loss": 0.012188894674181938,
            "image_validation_mae_degrees": 9.110383987426758,
            "image_validation_p95_degrees": 26.980209350585938,
            "wall_seconds": 23.606437100010226
          },
          {
            "step": 12000,
            "loss": 0.011083418503403664,
            "image_validation_mae_degrees": 9.02816390991211,
            "image_validation_p95_degrees": 26.53139877319336,
            "wall_seconds": 25.659855000005336
          },
          {
            "step": 13000,
            "loss": 0.009605573490262032,
            "image_validation_mae_degrees": 8.87343692779541,
            "image_validation_p95_degrees": 26.322805404663086,
            "wall_seconds": 27.82518469999195
          },
          {
            "step": 14000,
            "loss": 0.009236712008714676,
            "image_validation_mae_degrees": 8.8617525100708,
            "image_validation_p95_degrees": 26.245824813842773,
            "wall_seconds": 29.994265699991956
          },
          {
            "step": 15000,
            "loss": 0.008893096819519997,
            "image_validation_mae_degrees": 8.759113311767578,
            "image_validation_p95_degrees": 25.78264617919922,
            "wall_seconds": 31.995801400014898
          },
          {
            "step": 16000,
            "loss": 0.008174695074558258,
            "image_validation_mae_degrees": 8.756553649902344,
            "image_validation_p95_degrees": 26.338640213012695,
            "wall_seconds": 34.031909900018945
          },
          {
            "step": 17000,
            "loss": 0.007131262682378292,
            "image_validation_mae_degrees": 8.621706008911133,
            "image_validation_p95_degrees": 25.614479064941406,
            "wall_seconds": 36.128612600005
          },
          {
            "step": 18000,
            "loss": 0.006871702149510384,
            "image_validation_mae_degrees": 8.687061309814453,
            "image_validation_p95_degrees": 25.644140243530273,
            "wall_seconds": 38.245128900016425
          },
          {
            "step": 19000,
            "loss": 0.006526781711727381,
            "image_validation_mae_degrees": 8.54904556274414,
            "image_validation_p95_degrees": 25.678468704223633,
            "wall_seconds": 40.26716459999443
          },
          {
            "step": 20000,
            "loss": 0.0060137491673231125,
            "image_validation_mae_degrees": 8.516338348388672,
            "image_validation_p95_degrees": 25.40965461730957,
            "wall_seconds": 42.37103740000748
          },
          {
            "step": 21000,
            "loss": 0.005812383722513914,
            "image_validation_mae_degrees": 8.504302024841309,
            "image_validation_p95_degrees": 25.45140838623047,
            "wall_seconds": 44.36457910001627
          },
          {
            "step": 22000,
            "loss": 0.005753318313509226,
            "image_validation_mae_degrees": 8.423351287841797,
            "image_validation_p95_degrees": 25.26524543762207,
            "wall_seconds": 46.49428070001886
          },
          {
            "step": 23000,
            "loss": 0.005402491427958012,
            "image_validation_mae_degrees": 8.41252613067627,
            "image_validation_p95_degrees": 25.208173751831055,
            "wall_seconds": 48.641251799999736
          },
          {
            "step": 24000,
            "loss": 0.0057061645202338696,
            "image_validation_mae_degrees": 8.410500526428223,
            "image_validation_p95_degrees": 25.373044967651367,
            "wall_seconds": 50.67397840000922
          }
        ]
      },
      "outputs/online/training-v6-broad-w512.json": {
        "sha256": "7606f452348858e07b4f7f5978230add6c2a62c834b96b9bde7806d1c0de93e4",
        "metadata": {
          "method": "Supervised teacher examples; DAgger when aggregated datasets are supplied",
          "inputs": "Fresh depth image, continuous current XYZ and yaw, continuous goal XYZ",
          "parameters": 503554,
          "width": 512,
          "training_examples": 950000,
          "validation_examples": 50000,
          "steps": 24000,
          "batch": 2048,
          "seed": 130941,
          "data": [
            "outputs/online/imitation-initial.npz",
            "outputs/online/imitation-broad.npz"
          ],
          "resume": null,
          "learning_rate": 0.001,
          "inference": "Fresh network forward each 100 ms; no map-output cache, no route planner, no NAV IDs in policy",
          "validation_scope": "Image split is a fitting diagnostic only; navigation validation uses separate development situations."
        },
        "trace": [
          {
            "step": 1000,
            "loss": 0.19873858988285065,
            "image_validation_mae_degrees": 32.088008880615234,
            "image_validation_p95_degrees": 92.36779022216797,
            "wall_seconds": 2.8051674000162166
          },
          {
            "step": 2000,
            "loss": 0.17701971530914307,
            "image_validation_mae_degrees": 22.469524383544922,
            "image_validation_p95_degrees": 63.80668258666992,
            "wall_seconds": 5.080983699997887
          },
          {
            "step": 3000,
            "loss": 0.15315069258213043,
            "image_validation_mae_degrees": 17.360553741455078,
            "image_validation_p95_degrees": 50.11725616455078,
            "wall_seconds": 7.434978500008583
          },
          {
            "step": 4000,
            "loss": 0.12492300570011139,
            "image_validation_mae_degrees": 14.810020446777344,
            "image_validation_p95_degrees": 42.55668258666992,
            "wall_seconds": 9.755328500003088
          },
          {
            "step": 5000,
            "loss": 0.10378976166248322,
            "image_validation_mae_degrees": 13.44974136352539,
            "image_validation_p95_degrees": 38.158905029296875,
            "wall_seconds": 11.985548500000732
          },
          {
            "step": 6000,
            "loss": 0.09090764820575714,
            "image_validation_mae_degrees": 12.530813217163086,
            "image_validation_p95_degrees": 35.63861846923828,
            "wall_seconds": 14.20736030000262
          },
          {
            "step": 7000,
            "loss": 0.07826805114746094,
            "image_validation_mae_degrees": 12.008252143859863,
            "image_validation_p95_degrees": 33.97856521606445,
            "wall_seconds": 16.51717240002472
          },
          {
            "step": 8000,
            "loss": 0.07228168845176697,
            "image_validation_mae_degrees": 11.633162498474121,
            "image_validation_p95_degrees": 33.19248580932617,
            "wall_seconds": 18.753160000022035
          },
          {
            "step": 9000,
            "loss": 0.057709209620952606,
            "image_validation_mae_degrees": 11.0974760055542,
            "image_validation_p95_degrees": 31.419532775878906,
            "wall_seconds": 20.929133500001626
          },
          {
            "step": 10000,
            "loss": 0.05493490770459175,
            "image_validation_mae_degrees": 10.98245620727539,
            "image_validation_p95_degrees": 31.54029655456543,
            "wall_seconds": 23.249104000016814
          },
          {
            "step": 11000,
            "loss": 0.05018517002463341,
            "image_validation_mae_degrees": 10.712691307067871,
            "image_validation_p95_degrees": 30.367679595947266,
            "wall_seconds": 25.598264200001722
          },
          {
            "step": 12000,
            "loss": 0.04601769149303436,
            "image_validation_mae_degrees": 10.60275936126709,
            "image_validation_p95_degrees": 30.277971267700195,
            "wall_seconds": 27.71912120000343
          },
          {
            "step": 13000,
            "loss": 0.042176004499197006,
            "image_validation_mae_degrees": 10.40385913848877,
            "image_validation_p95_degrees": 29.734338760375977,
            "wall_seconds": 29.920460900000762
          },
          {
            "step": 14000,
            "loss": 0.037423159927129745,
            "image_validation_mae_degrees": 10.415765762329102,
            "image_validation_p95_degrees": 29.537952423095703,
            "wall_seconds": 32.19032670001616
          },
          {
            "step": 15000,
            "loss": 0.03595757856965065,
            "image_validation_mae_degrees": 10.177858352661133,
            "image_validation_p95_degrees": 28.869417190551758,
            "wall_seconds": 34.39035200001672
          },
          {
            "step": 16000,
            "loss": 0.03619365766644478,
            "image_validation_mae_degrees": 10.099966049194336,
            "image_validation_p95_degrees": 29.02203941345215,
            "wall_seconds": 36.578681800019694
          },
          {
            "step": 17000,
            "loss": 0.03280169144272804,
            "image_validation_mae_degrees": 9.985811233520508,
            "image_validation_p95_degrees": 28.46563148498535,
            "wall_seconds": 38.919168100022944
          },
          {
            "step": 18000,
            "loss": 0.029290683567523956,
            "image_validation_mae_degrees": 9.930652618408203,
            "image_validation_p95_degrees": 28.350385665893555,
            "wall_seconds": 41.24041150000994
          },
          {
            "step": 19000,
            "loss": 0.027591364458203316,
            "image_validation_mae_degrees": 9.834296226501465,
            "image_validation_p95_degrees": 28.273874282836914,
            "wall_seconds": 43.37046689999988
          },
          {
            "step": 20000,
            "loss": 0.027947863563895226,
            "image_validation_mae_degrees": 9.833388328552246,
            "image_validation_p95_degrees": 28.31706428527832,
            "wall_seconds": 45.758660799998324
          },
          {
            "step": 21000,
            "loss": 0.026641592383384705,
            "image_validation_mae_degrees": 9.755372047424316,
            "image_validation_p95_degrees": 27.80673599243164,
            "wall_seconds": 48.135226900019916
          },
          {
            "step": 22000,
            "loss": 0.025031551718711853,
            "image_validation_mae_degrees": 9.720199584960938,
            "image_validation_p95_degrees": 28.019636154174805,
            "wall_seconds": 50.3997803000093
          },
          {
            "step": 23000,
            "loss": 0.025851961225271225,
            "image_validation_mae_degrees": 9.668376922607422,
            "image_validation_p95_degrees": 27.76698875427246,
            "wall_seconds": 52.63903360001859
          },
          {
            "step": 24000,
            "loss": 0.023374035954475403,
            "image_validation_mae_degrees": 9.606120109558105,
            "image_validation_p95_degrees": 27.761768341064453,
            "wall_seconds": 54.96148370002629
          }
        ]
      },
      "outputs/online/training-v7-broad-r1.json": {
        "sha256": "b9c58600f64b839dac9e1374813ddb009b04bd5b1fed1bee0a30be2f603384f9",
        "metadata": {
          "method": "Supervised teacher examples; DAgger when aggregated datasets are supplied",
          "inputs": "Fresh depth image, continuous current XYZ and yaw, continuous goal XYZ",
          "parameters": 1793538,
          "width": 1024,
          "dagger_sampling_fraction": 0.5,
          "training_examples": 986874,
          "validation_examples": 51941,
          "steps": 8000,
          "batch": 2048,
          "seed": 130941,
          "data": [
            "outputs/online/imitation-initial.npz",
            "outputs/online/imitation-broad.npz",
            "outputs/online/dagger-broad-r1.npz"
          ],
          "resume": "outputs/online/checkpoint/v6-selected-development.pt",
          "learning_rate": 0.0003,
          "inference": "Fresh network forward each 100 ms; no map-output cache, no route planner, no NAV IDs in policy",
          "validation_scope": "Image split is a fitting diagnostic only; navigation validation uses separate development situations."
        },
        "trace": [
          {
            "step": 1000,
            "loss": 0.007346144877374172,
            "image_validation_mae_degrees": 8.532886505126953,
            "image_validation_p95_degrees": 25.742223739624023,
            "wall_seconds": 2.569004099990707
          },
          {
            "step": 2000,
            "loss": 0.01242522057145834,
            "image_validation_mae_degrees": 8.822456359863281,
            "image_validation_p95_degrees": 26.05653190612793,
            "wall_seconds": 4.841267699986929
          },
          {
            "step": 3000,
            "loss": 0.009201344102621078,
            "image_validation_mae_degrees": 8.650151252746582,
            "image_validation_p95_degrees": 25.942829132080078,
            "wall_seconds": 7.298894699983066
          },
          {
            "step": 4000,
            "loss": 0.00877933669835329,
            "image_validation_mae_degrees": 8.682160377502441,
            "image_validation_p95_degrees": 25.746591567993164,
            "wall_seconds": 9.727798599982634
          },
          {
            "step": 5000,
            "loss": 0.0074692764319479465,
            "image_validation_mae_degrees": 8.64272689819336,
            "image_validation_p95_degrees": 25.81437110900879,
            "wall_seconds": 12.369010399997933
          },
          {
            "step": 6000,
            "loss": 0.0070882984437048435,
            "image_validation_mae_degrees": 8.631049156188965,
            "image_validation_p95_degrees": 25.624271392822266,
            "wall_seconds": 15.605406499991659
          },
          {
            "step": 7000,
            "loss": 0.0065001267939805984,
            "image_validation_mae_degrees": 8.564538955688477,
            "image_validation_p95_degrees": 25.4715633392334,
            "wall_seconds": 18.822251599980518
          },
          {
            "step": 8000,
            "loss": 0.00673706317320466,
            "image_validation_mae_degrees": 8.585027694702148,
            "image_validation_p95_degrees": 25.706289291381836,
            "wall_seconds": 22.012822199991206
          }
        ]
      },
      "outputs/online/training-v7-broad-r2.json": {
        "sha256": "0ca98e18bd3f28106adef2925e222ae703d34ef0165ffd0befc79711ba43d5d9",
        "metadata": {
          "method": "Supervised teacher examples; DAgger when aggregated datasets are supplied",
          "inputs": "Fresh depth image, continuous current XYZ and yaw, continuous goal XYZ",
          "parameters": 1793538,
          "width": 1024,
          "dagger_sampling_fraction": 0.5,
          "training_examples": 1020300,
          "validation_examples": 53700,
          "steps": 8000,
          "batch": 2048,
          "seed": 130941,
          "data": [
            "outputs/online/imitation-initial.npz",
            "outputs/online/imitation-broad.npz",
            "outputs/online/dagger-broad-r1.npz",
            "outputs/online/dagger-broad-r2.npz"
          ],
          "resume": "outputs/online/checkpoint/v7-broad-r1-04000.pt",
          "learning_rate": 0.00015,
          "inference": "Fresh network forward each 100 ms; no map-output cache, no route planner, no NAV IDs in policy",
          "validation_scope": "Image split is a fitting diagnostic only; navigation validation uses separate development situations."
        },
        "trace": [
          {
            "step": 1000,
            "loss": 0.005842684768140316,
            "image_validation_mae_degrees": 6.623046875,
            "image_validation_p95_degrees": 19.044143676757812,
            "wall_seconds": 2.675039099995047
          },
          {
            "step": 2000,
            "loss": 0.010599943809211254,
            "image_validation_mae_degrees": 6.901649475097656,
            "image_validation_p95_degrees": 19.29251480102539,
            "wall_seconds": 4.755460199987283
          },
          {
            "step": 3000,
            "loss": 0.009900854900479317,
            "image_validation_mae_degrees": 6.874927043914795,
            "image_validation_p95_degrees": 19.585142135620117,
            "wall_seconds": 6.861475499987137
          },
          {
            "step": 4000,
            "loss": 0.00873463787138462,
            "image_validation_mae_degrees": 6.964544773101807,
            "image_validation_p95_degrees": 19.783357620239258,
            "wall_seconds": 9.09533250000095
          },
          {
            "step": 5000,
            "loss": 0.0077987778931856155,
            "image_validation_mae_degrees": 6.923450946807861,
            "image_validation_p95_degrees": 19.706188201904297,
            "wall_seconds": 11.339572199998656
          },
          {
            "step": 6000,
            "loss": 0.0070532336831092834,
            "image_validation_mae_degrees": 7.028418064117432,
            "image_validation_p95_degrees": 19.964181900024414,
            "wall_seconds": 13.459194099996239
          },
          {
            "step": 7000,
            "loss": 0.006848536431789398,
            "image_validation_mae_degrees": 6.981973648071289,
            "image_validation_p95_degrees": 19.873876571655273,
            "wall_seconds": 15.5839999000018
          },
          {
            "step": 8000,
            "loss": 0.006459639873355627,
            "image_validation_mae_degrees": 7.051494121551514,
            "image_validation_p95_degrees": 20.198638916015625,
            "wall_seconds": 17.79650089997449
          }
        ]
      },
      "outputs/online/training-v7-broad-r3.json": {
        "sha256": "94c9c4c2640bc9350f06295750c92e5deb782bf9db5423f7d9c537ce535138d1",
        "metadata": {
          "method": "Supervised teacher examples; DAgger when aggregated datasets are supplied",
          "inputs": "Fresh depth image, continuous current XYZ and yaw, continuous goal XYZ",
          "parameters": 1793538,
          "width": 1024,
          "dagger_sampling_fraction": 0.5,
          "training_examples": 1051878,
          "validation_examples": 55363,
          "steps": 8000,
          "batch": 2048,
          "seed": 130941,
          "data": [
            "outputs/online/imitation-initial.npz",
            "outputs/online/imitation-broad.npz",
            "outputs/online/dagger-broad-r1.npz",
            "outputs/online/dagger-broad-r2.npz",
            "outputs/online/dagger-broad-r3.npz"
          ],
          "resume": "outputs/online/checkpoint/v7-broad-r2-08000.pt",
          "learning_rate": 0.0001,
          "inference": "Fresh network forward each 100 ms; no map-output cache, no route planner, no NAV IDs in policy",
          "validation_scope": "Image split is a fitting diagnostic only; navigation validation uses separate development situations."
        },
        "trace": [
          {
            "step": 1000,
            "loss": 0.005025646183639765,
            "image_validation_mae_degrees": 6.300022125244141,
            "image_validation_p95_degrees": 18.338298797607422,
            "wall_seconds": 2.7343402999977116
          },
          {
            "step": 2000,
            "loss": 0.007932199165225029,
            "image_validation_mae_degrees": 6.494313716888428,
            "image_validation_p95_degrees": 18.594072341918945,
            "wall_seconds": 5.017455900000641
          },
          {
            "step": 3000,
            "loss": 0.007303212303668261,
            "image_validation_mae_degrees": 6.485983848571777,
            "image_validation_p95_degrees": 18.785552978515625,
            "wall_seconds": 7.118603700015228
          },
          {
            "step": 4000,
            "loss": 0.006110438145697117,
            "image_validation_mae_degrees": 6.556896209716797,
            "image_validation_p95_degrees": 18.82453155517578,
            "wall_seconds": 9.236401000001933
          },
          {
            "step": 5000,
            "loss": 0.005811783485114574,
            "image_validation_mae_degrees": 6.556222438812256,
            "image_validation_p95_degrees": 18.946144104003906,
            "wall_seconds": 11.464619400008814
          },
          {
            "step": 6000,
            "loss": 0.005863398313522339,
            "image_validation_mae_degrees": 6.584855556488037,
            "image_validation_p95_degrees": 18.98244285583496,
            "wall_seconds": 13.694784899998922
          },
          {
            "step": 7000,
            "loss": 0.005533978343009949,
            "image_validation_mae_degrees": 6.579241752624512,
            "image_validation_p95_degrees": 18.993610382080078,
            "wall_seconds": 15.831587099994067
          },
          {
            "step": 8000,
            "loss": 0.0057338448241353035,
            "image_validation_mae_degrees": 6.6057939529418945,
            "image_validation_p95_degrees": 19.048742294311523,
            "wall_seconds": 17.979172899998957
          }
        ]
      }
    },
    "datasets": {
      "outputs/online/dagger-broad-r1.npz": {
        "examples": 38815,
        "features": 212,
        "target_dimensions": 2,
        "sha256": "e83a235261404947a3ee7a70b6ba093df3d1e81f4165d86d755b0e0c98a4475d"
      },
      "outputs/online/dagger-broad-r2.npz": {
        "examples": 35185,
        "features": 212,
        "target_dimensions": 2,
        "sha256": "4fa00a60067615c7397135086b90f8c28f22016e2a92586f39872d72aaf44c7c"
      },
      "outputs/online/dagger-broad-r3.npz": {
        "examples": 33241,
        "features": 212,
        "target_dimensions": 2,
        "sha256": "88cef0c364ded6797190da1a2a298b09eff627c072d45dce360c169d4bad9872"
      },
      "outputs/online/imitation-broad.npz": {
        "examples": 900000,
        "features": 212,
        "target_dimensions": 2,
        "sha256": "eec51b43f7b949f9a5f7a53024d29517529924e256f6a952811e59dc94167516"
      },
      "outputs/online/imitation-initial.npz": {
        "examples": 100000,
        "features": 212,
        "target_dimensions": 2,
        "sha256": "885ee6d20d3104e461feb158fb54d25ff24aa4574f1b92c5523b0a1f5aa0bc68"
      }
    },
    "discarded_pilots": [
      {
        "report": "v6-development-pilot.json",
        "successes": 1,
        "trials": 8,
        "selected": false
      },
      {
        "report": "v7-development-pilot-dagger.json",
        "successes": 0,
        "trials": 8,
        "selected": false
      },
      {
        "report": "v8-development-pilot-050.json",
        "successes": 0,
        "trials": 8,
        "selected": false
      }
    ],
    "discarded_rl_pilot": {
      "status": "discarded development pilot",
      "tag": "rl-scratch",
      "reason": "Original process rendered images on AMD integrated GPU. Restarted PPO from scratch on project-local NVIDIA launcher for a larger training budget.",
      "transferred_weights": false
    },
    "compute": {
      "python": "3.13.15",
      "pytorch": "2.14.0+cu130",
      "cuda": "13.0",
      "device": "NVIDIA GeForce RTX 5090",
      "launcher": "python-gpu.exe",
      "depth_renderer": "Original single-view DepthRenderer on project-local NVIDIA graphics configuration",
      "timing_note": "Per-run timings include contention from concurrent GPU jobs. Dataset-generation time is recorded separately."
    },
    "data_generation": {
      "count": 900000,
      "seed": 130943,
      "features": 212,
      "generation_seconds": 2076.2584685000184,
      "teacher": "16-step local Dijkstra portal teacher, training only; numerically checked against GoalNavigator on 64 training states"
    },
    "scope": "Training and development evidence. Final navigation results are generated only after freezing these files."
  },
  "changed_target": {
    "scope": "Eight fixed development inputs; controlled one-step probes, not held-out navigation success",
    "renderer": "NVIDIA GeForce RTX 5090/PCIe/SSE2",
    "models": {
      "v6": {
        "sensory_parameters": 1793538,
        "probes": {
          "normal": {
            "cue": [
              0.49737058622009833,
              -0.8708457926002073,
              0.44708048277258605,
              3.3573217043400527,
              -0.15833831257215625,
              0.11158633726472544,
              -0.10555207938433009,
              -1.3169606750240337
            ],
            "speed": [
              0.013602163642644882,
              2.5248165130615234,
              0.003742220578715205,
              1.792125940322876,
              0.0062350803054869175,
              0.181813046336174,
              0.009787237271666527,
              3.1548209190368652
            ],
            "turn": [
              2.799999952316284,
              0.6934909224510193,
              2.799999952316284,
              1.1048095226287842,
              2.799999952316284,
              2.799999952316284,
              2.799999952316284,
              -0.42617475986480713
            ]
          },
          "no_image": {
            "cue": [
              0.3199061010516413,
              0.116416411689229,
              0.40036999159216613,
              2.9652083540082694,
              -5.589405540889417,
              -0.270013565841964,
              -0.15014732093096095,
              -1.0254088764419567
            ],
            "speed": [
              0.0230413805693388,
              0.22520162165164948,
              0.004064230713993311,
              4.1683349609375,
              0.00484623946249485,
              0.38081276416778564,
              0.010956984013319016,
              2.8110029697418213
            ],
            "turn": [
              2.799999952316284,
              2.799999952316284,
              2.799999952316284,
              -0.09187285602092743,
              -2.799999952316284,
              2.799999952316284,
              2.799999952316284,
              0.5646433234214783
            ]
          },
          "changed_target": {
            "cue": [
              0.47619424802429755,
              -0.7231675812688398,
              0.2642098563230011,
              0.7660302171231033,
              -0.04513335651746875,
              1.4659187843167274,
              -0.12451923102617579,
              -1.4366469954481136
            ],
            "speed": [
              0.014303362928330898,
              1.6841930150985718,
              0.00523090735077858,
              0.01611369661986828,
              0.004768033046275377,
              0.005186491645872593,
              0.010281214490532875,
              2.3029885292053223
            ],
            "turn": [
              2.799999952316284,
              1.1793490648269653,
              2.799999952316284,
              -2.799999952316284,
              2.799999952316284,
              2.799999952316284,
              2.799999952316284,
              -0.8038437366485596
            ]
          }
        },
        "effects": {
          "no_image": {
            "mean_cue_change_degrees": 22.727914728994996,
            "max_cue_change_degrees": 56.56595757856718,
            "mean_motor_turn_change_rad_s": 1.2367511745542288
          },
          "changed_target": {
            "mean_cue_change_degrees": 32.58129900968606,
            "max_cue_change_degrees": 148.47006570570952,
            "mean_motor_turn_change_rad_s": 0.5960420742630959
          }
        },
        "zero_PFL3_stop": true,
        "policy_forward_calls": 3,
        "full_graph_neurons": 166700,
        "full_graph_edges": 25582938,
        "cache_or_teacher_loaded": false
      },
      "v7": {
        "sensory_parameters": 1793538,
        "probes": {
          "normal": {
            "cue": [
              0.5197197054064997,
              -0.6946777233567762,
              0.5850351946866486,
              3.3792487630367996,
              -0.1017823261677373,
              0.13348997133607554,
              -0.1468578597950967,
              -1.184361245796622
            ],
            "speed": [
              0.012949108146131039,
              1.5440415143966675,
              0.0027813285123556852,
              1.6795248985290527,
              0.00549342343583703,
              0.1800992488861084,
              0.010869933292269707,
              4.011160850524902
            ],
            "turn": [
              2.799999952316284,
              1.2836087942123413,
              2.799999952316284,
              1.1826797723770142,
              2.799999952316284,
              2.799999952316284,
              2.799999952316284,
              0.13799601793289185
            ]
          },
          "no_image": {
            "cue": [
              0.6238595102465876,
              -0.05867580861716748,
              -4.593470941253903,
              2.806149984312129,
              -5.249612812465345,
              -0.30444359285002065,
              -0.059777857964042,
              1.4131182009944905
            ],
            "speed": [
              0.010604150593280792,
              0.2798812985420227,
              0.018605070188641548,
              2.519070863723755,
              0.013280411250889301,
              0.4169890284538269,
              0.008600383065640926,
              0.007461527828127146
            ],
            "turn": [
              2.799999952316284,
              2.799999952316284,
              -2.799999952316284,
              -0.6962246894836426,
              -2.799999952316284,
              2.799999952316284,
              2.799999952316284,
              2.799999952316284
            ]
          },
          "changed_target": {
            "cue": [
              0.5778652284777888,
              -0.7629459926572371,
              0.4223970072782013,
              0.6781157502651931,
              0.1191356139720332,
              1.2823834468686073,
              -0.07184422225237208,
              -1.4327791159381877
            ],
            "speed": [
              0.011562449857592583,
              1.8986073732376099,
              0.003908019978553057,
              0.011057647876441479,
              0.0024964106269180775,
              0.00934998132288456,
              0.008905638009309769,
              2.3115508556365967
            ],
            "turn": [
              2.799999952316284,
              1.0355479717254639,
              2.799999952316284,
              -2.799999952316284,
              2.799999952316284,
              2.799999952316284,
              2.799999952316284,
              -0.799390435218811
            ]
          }
        },
        "effects": {
          "no_image": {
            "mean_cue_change_degrees": 47.811662868234365,
            "max_cue_change_degrees": 148.82460967310664,
            "mean_motor_turn_change_rad_s": 2.157162420451641
          },
          "changed_target": {
            "mean_cue_change_degrees": 33.54258165441277,
            "max_cue_change_degrees": 154.76352153526975,
            "mean_motor_turn_change_rad_s": 0.6460158750414848
          }
        },
        "zero_PFL3_stop": true,
        "policy_forward_calls": 3,
        "full_graph_neurons": 166700,
        "full_graph_edges": 25582938,
        "cache_or_teacher_loaded": false
      },
      "v8": {
        "sensory_parameters": 120577,
        "probes": {
          "normal": {
            "cue": [
              0.9961089821437719,
              -1.2455196594288511,
              1.0484971108541647,
              -0.4813229882898751,
              0.5144071316435828,
              1.026828508576955,
              0.6959417352234747,
              2.407148865295408
            ],
            "speed": [
              2.9660134315490723,
              3.1569299697875977,
              3.090623140335083,
              3.628972291946411,
              2.9191460609436035,
              2.915846824645996,
              2.937878131866455,
              3.970738410949707
            ],
            "turn": [
              0.026059910655021667,
              0.11311545222997665,
              -0.5290096998214722,
              0.7126439809799194,
              0.009184920229017735,
              0.49650728702545166,
              -0.04810425639152527,
              -0.2234007716178894
            ]
          },
          "no_image": {
            "cue": [
              1.3949929432848813,
              -1.3113435124447508,
              1.6204618516073386,
              -0.08763442875770955,
              2.0057552790358555,
              0.70128127535304,
              1.2190186271225836,
              2.7517660160193422
            ],
            "speed": [
              3.504291296005249,
              3.127436876296997,
              3.741236686706543,
              3.0411062240600586,
              3.7389237880706787,
              2.582465887069702,
              3.2357640266418457,
              3.8134958744049072
            ],
            "turn": [
              -0.3658672869205475,
              -0.007308363914489746,
              -1.0810153484344482,
              0.5176034569740295,
              -1.2379530668258667,
              0.7959155440330505,
              -0.825274646282196,
              -0.3371328115463257
            ]
          },
          "changed_target": {
            "cue": [
              0.986959272576033,
              0.9000893192082666,
              0.09068695358533885,
              1.4467578922997055,
              -0.7085867203166051,
              1.4347445830477024,
              1.3729999823924928,
              -1.728910310162682
            ],
            "speed": [
              2.9541263580322266,
              2.6496639251708984,
              2.8977742195129395,
              3.3108677864074707,
              3.9006412029266357,
              3.468543291091919,
              3.394287586212158,
              3.1576826572418213
            ],
            "turn": [
              0.03501246124505997,
              0.6517150402069092,
              0.809938907623291,
              -1.142024040222168,
              1.1155500411987305,
              0.07500575482845306,
              -0.9806782007217407,
              -0.3733711242675781
            ]
          }
        },
        "effects": {
          "no_image": {
            "mean_cue_change_degrees": 29.47116233928041,
            "max_cue_change_degrees": 85.44795463023145,
            "mean_motor_turn_change_rad_s": 0.4621057325275615
          },
          "changed_target": {
            "mean_cue_change_degrees": 68.00825704186076,
            "max_cue_change_degrees": 123.02126542989222,
            "mean_motor_turn_change_rad_s": 0.7939474646700546
          }
        },
        "zero_PFL3_stop": true,
        "policy_forward_calls": 3,
        "full_graph_neurons": 166700,
        "full_graph_edges": 25582938,
        "cache_or_teacher_loaded": false
      }
    },
    "parameters": {
      "v6_visual": 1793538,
      "v7_visual": 1793538,
      "v8_sensory": 120577,
      "v8_motor": 5890,
      "v8_critic_training_only": 120577
    }
  }
}