{
 "arm": "lev-4b",
 "card": "RTX 3090 (fp 33a0b4acb8e8)",
 "card_empty_after": true,
 "driver": [
  "595.84",
  "NVIDIA GeForce RTX 3090",
  "24576"
 ],
 "dry": 0,
 "ended": "2026-09-28T13:19:40Z",
 "env": {
  "DO_NOT_TRACK": null,
  "GRADIO_ANALYTICS_ENABLED": "False",
  "HF_HOME": "/workshop/bench-scratch-lev2/sandbox-home/hf",
  "HF_HUB_DISABLE_TELEMETRY": "1",
  "HF_HUB_OFFLINE": "1",
  "ORT_DISABLE_TELEMETRY": "1",
  "TORCH_DISABLE_NATIVE_JIT": "1",
  "TRANSFORMERS_OFFLINE": "1",
  "VLLM_NO_USAGE_STATS": "1"
 },
 "first_row_task": "kc",
 "health": {
  "calibrated": true,
  "checkpoint": "/workshop/bench-scratch-lev2/weights/lev-f8ef7115",
  "compiled": false,
  "max_label_options": null,
  "mode_b": true,
  "model": "Qwen/Qwen3.5-4B",
  "noul_readout": "rating",
  "order_average": true,
  "prefix_mode": "single",
  "prompt_style": "chat",
  "skip_multi_token_codes": true,
  "status": "ok"
 },
 "hostos_sha256": "a6a74bba57250be1936825466c6614876d757cf81373fcf6fa0b48587cf0e3ad",
 "instrument_sha256": {
  "run.py": "<withheld: rewritten file, see README.md>",
  "run_addenda.py": "a34aca61833ed0c5860c91c7e2e227429e8ea9e2e50fec68d7b7fec15d89acab",
  "run_kev.py": "1fbce7a41746ed3d612a8a488e35dbebba789bc55086daa7d71ffce2cd010b05"
 },
 "os": {
  "arch": "x86_64",
  "id": "ubuntu",
  "kernel": "7.0.0-31-generic",
  "pretty_name": "Ubuntu 26.04 LTS",
  "source": "/etc/os-release",
  "version_id": "26.04"
 },
 "posture_after": {
  "compute_apps": 0,
  "compute_apps_mib": [],
  "memory_used_mib": 1.0,
  "persistence_read_inside_sandbox_not_of_record": "Disabled",
  "posture_ok": true,
  "power_limit_w": 300.0,
  "pstate": "P3",
  "temp_c": 65.0,
  "utc": "2026-09-28T13:19:40Z",
  "when": "after the server exited"
 },
 "posture_after_tasks": {
  "compute_apps": 1,
  "compute_apps_mib": [
   22726.0
  ],
  "memory_used_mib": 22736.0,
  "persistence_read_inside_sandbox_not_of_record": "Disabled",
  "posture_ok": true,
  "power_limit_w": 300.0,
  "pstate": "P2",
  "temp_c": 67.0,
  "utc": "2026-09-28T13:19:37Z",
  "when": "after the tasks, server still up"
 },
 "posture_before": {
  "compute_apps": 0,
  "compute_apps_mib": [],
  "memory_used_mib": 1.0,
  "persistence_read_inside_sandbox_not_of_record": "Disabled",
  "posture_ok": true,
  "power_limit_w": 300.0,
  "pstate": "P8",
  "temp_c": 33.0,
  "utc": "2026-09-28T13:12:27Z",
  "when": "before the server"
 },
 "posture_resident": {
  "compute_apps": 1,
  "compute_apps_mib": [
   9116.0
  ],
  "memory_used_mib": 9126.0,
  "persistence_read_inside_sandbox_not_of_record": "Disabled",
  "posture_ok": true,
  "power_limit_w": 300.0,
  "pstate": "P2",
  "temp_c": 38.0,
  "utc": "2026-09-28T13:12:44Z",
  "when": "after the warm-ups"
 },
 "residency": {
  "bar_mib": 7788.0,
  "compute_apps": 1,
  "growth_mib": 9125.0,
  "ok": true
 },
 "runner_sha256": "e2f026dae167816a26305373e9b16ad9ec7ca197c7dc08530c56f608bc3df0f0",
 "server_ready_s": 7.0,
 "started": "2026-09-28T13:12:27Z",
 "summaries": {
  "ka": {
   "accuracy": 0.9811320754716981,
   "brier": 0.031143057812033397,
   "correct": 52,
   "energy": {
    "cards": [
     "GPU-fp:33a0b4acb8e8"
    ],
    "decisions": 53,
    "joules": 31698.823408539294,
    "joules_per_decision": 598.0910077082885,
    "mean_watts": 296.1842524054565,
    "net_joules_per_decision": 338.8358508397565,
    "peak_watts": 300.06,
    "per_card": {
     "GPU-fp:33a0b4acb8e8": {
      "joules": 31698.823408539294,
      "mean_watts": 296.1842524054565,
      "min_watts": 111.18,
      "peak_watts": 300.06,
      "samples": 108,
      "seconds": 107.0239999294281
     }
    },
    "samples": 108,
    "seconds": 107.0239999294281
   },
   "errors": 0,
   "eval_tokens_total": 0,
   "floored_rows": 0,
   "idle": {
    "cards": [
     "GPU-fp:33a0b4acb8e8"
    ],
    "joules": 1156.7696498918533,
    "mean_watts": 128.38730866995007,
    "peak_watts": 260.22,
    "per_card": {
     "GPU-fp:33a0b4acb8e8": {
      "joules": 1156.7696498918533,
      "mean_watts": 128.38730866995007,
      "min_watts": 111.15,
      "peak_watts": 260.22,
      "samples": 10,
      "seconds": 9.009999990463257
     }
    },
    "samples": 10,
    "seconds": 9.009999990463257,
    "what": "the draw of these cards immediately before the arm, with the model resident and no work in flight"
   },
   "letters_floored_total": 0,
   "median_seconds": 1.7306802479997714,
   "n": 63,
   "n_labelled": 53,
   "n_reached": 53,
   "p95_seconds": 3.615360237999994,
   "prompt_tokens_total": 376622,
   "refused": 10,
   "refused_by_reason": {
    "overlong": 10
   },
   "reliability": [
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.1,
     "lo": 0.0,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.2,
     "lo": 0.1,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.3,
     "lo": 0.2,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.4,
     "lo": 0.3,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.5,
     "lo": 0.4,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": 0.5,
     "correct": 1,
     "gap": 0.01973277547813379,
     "hi": 0.6,
     "lo": 0.5,
     "mean_p": 0.5197327754781338,
     "n": 2,
     "sum_p": 1.0394655509562676
    },
    {
     "accuracy": 1.0,
     "correct": 1,
     "gap": -0.34079189071578875,
     "hi": 0.7,
     "lo": 0.6,
     "mean_p": 0.6592081092842113,
     "n": 1,
     "sum_p": 0.6592081092842113
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.8,
     "lo": 0.7,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": 1.0,
     "correct": 4,
     "gap": -0.13770476721086133,
     "hi": 0.9,
     "lo": 0.8,
     "mean_p": 0.8622952327891387,
     "n": 4,
     "sum_p": 3.4491809311565547
    },
    {
     "accuracy": 1.0,
     "correct": 46,
     "gap": -0.056193218500415365,
     "hi": 1.0,
     "lo": 0.9,
     "mean_p": 0.9438067814995846,
     "n": 46,
     "sum_p": 43.41511194898089
    }
   ],
   "unparsed": 0,
   "wall_seconds": 107.17456943999969
  },
  "kc": {
   "accuracy": 0.6111111111111112,
   "brier": 0.5401864684760486,
   "correct": 66,
   "energy": {
    "cards": [
     "GPU-fp:33a0b4acb8e8"
    ],
    "decisions": 108,
    "joules": 6352.719135297537,
    "joules_per_decision": 58.821473474977196,
    "mean_watts": 288.64187910080847,
    "net_joules_per_decision": 32.51826087132361,
    "peak_watts": 299.66,
    "per_card": {
     "GPU-fp:33a0b4acb8e8": {
      "joules": 6352.719135297537,
      "mean_watts": 288.64187910080847,
      "min_watts": 109.0,
      "peak_watts": 299.66,
      "samples": 23,
      "seconds": 22.009000062942505
     }
    },
    "samples": 23,
    "seconds": 22.009000062942505
   },
   "errors": 0,
   "eval_tokens_total": 0,
   "floored_rows": 0,
   "idle": {
    "cards": [
     "GPU-fp:33a0b4acb8e8"
    ],
    "joules": 1162.810190331936,
    "mean_watts": 129.072059297127,
    "peak_watts": 202.27,
    "per_card": {
     "GPU-fp:33a0b4acb8e8": {
      "joules": 1162.810190331936,
      "mean_watts": 129.072059297127,
      "min_watts": 108.9,
      "peak_watts": 202.27,
      "samples": 10,
      "seconds": 9.009000062942505
     }
    },
    "samples": 10,
    "seconds": 9.009000062942505,
    "what": "the draw of these cards immediately before the arm, with the model resident and no work in flight"
   },
   "letters_floored_total": 0,
   "median_seconds": 0.21668299049997586,
   "n": 108,
   "n_labelled": 108,
   "n_reached": 108,
   "p95_seconds": 0.2495980049998252,
   "prompt_tokens_total": 63578,
   "refused": 0,
   "refused_by_reason": {},
   "reliability": [
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.1,
     "lo": 0.0,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.2,
     "lo": 0.1,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.3,
     "lo": 0.2,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": 0.2,
     "correct": 1,
     "gap": 0.15864685605364082,
     "hi": 0.4,
     "lo": 0.3,
     "mean_p": 0.35864685605364083,
     "n": 5,
     "sum_p": 1.7932342802682042
    },
    {
     "accuracy": 0.23076923076923078,
     "correct": 3,
     "gap": 0.22165952813579248,
     "hi": 0.5,
     "lo": 0.4,
     "mean_p": 0.45242875890502326,
     "n": 13,
     "sum_p": 5.8815738657653025
    },
    {
     "accuracy": 0.6,
     "correct": 6,
     "gap": -0.03067738675275511,
     "hi": 0.6,
     "lo": 0.5,
     "mean_p": 0.5693226132472449,
     "n": 10,
     "sum_p": 5.693226132472448
    },
    {
     "accuracy": 0.25,
     "correct": 3,
     "gap": 0.39860477874718614,
     "hi": 0.7,
     "lo": 0.6,
     "mean_p": 0.6486047787471861,
     "n": 12,
     "sum_p": 7.783257344966234
    },
    {
     "accuracy": 0.5555555555555556,
     "correct": 5,
     "gap": 0.20864407834304366,
     "hi": 0.8,
     "lo": 0.7,
     "mean_p": 0.7641996338985992,
     "n": 9,
     "sum_p": 6.877796705087393
    },
    {
     "accuracy": 0.8125,
     "correct": 26,
     "gap": 0.0453301670260049,
     "hi": 0.9,
     "lo": 0.8,
     "mean_p": 0.8578301670260049,
     "n": 32,
     "sum_p": 27.450565344832157
    },
    {
     "accuracy": 0.8148148148148148,
     "correct": 22,
     "gap": 0.13125102793485321,
     "hi": 1.0,
     "lo": 0.9,
     "mean_p": 0.946065842749668,
     "n": 27,
     "sum_p": 25.543777754241034
    }
   ],
   "unparsed": 0,
   "wall_seconds": 22.82549264499994
  },
  "kd": {
   "accuracy": 0.6458333333333334,
   "brier": 0.4961192913563542,
   "correct": 31,
   "energy": {
    "cards": [
     "GPU-fp:33a0b4acb8e8"
    ],
    "decisions": 48,
    "joules": 2502.4284649515153,
    "joules_per_decision": 52.13392635315657,
    "mean_watts": 277.8315169790452,
    "net_joules_per_decision": 26.068225614825888,
    "peak_watts": 299.71,
    "per_card": {
     "GPU-fp:33a0b4acb8e8": {
      "joules": 2502.4284649515153,
      "mean_watts": 277.8315169790452,
      "min_watts": 123.84,
      "peak_watts": 299.71,
      "samples": 10,
      "seconds": 9.006999969482422
     }
    },
    "samples": 10,
    "seconds": 9.006999969482422
   },
   "errors": 0,
   "eval_tokens_total": 0,
   "floored_rows": 0,
   "idle": {
    "cards": [
     "GPU-fp:33a0b4acb8e8"
    ],
    "joules": 1251.1536354398727,
    "mean_watts": 138.9090307182236,
    "peak_watts": 200.01,
    "per_card": {
     "GPU-fp:33a0b4acb8e8": {
      "joules": 1251.1536354398727,
      "mean_watts": 138.9090307182236,
      "min_watts": 124.36,
      "peak_watts": 200.01,
      "samples": 10,
      "seconds": 9.006999969482422
     }
    },
    "samples": 10,
    "seconds": 9.006999969482422,
    "what": "the draw of these cards immediately before the arm, with the model resident and no work in flight"
   },
   "letters_floored_total": 0,
   "median_seconds": 0.1938230295008907,
   "n": 48,
   "n_labelled": 48,
   "n_reached": 48,
   "p95_seconds": 0.19543845999942278,
   "prompt_tokens_total": 27142,
   "refused": 0,
   "refused_by_reason": {},
   "reliability": [
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.1,
     "lo": 0.0,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.2,
     "lo": 0.1,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.3,
     "lo": 0.2,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.4,
     "lo": 0.3,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.5,
     "lo": 0.4,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": 0.25,
     "correct": 1,
     "gap": 0.2888910541496227,
     "hi": 0.6,
     "lo": 0.5,
     "mean_p": 0.5388910541496227,
     "n": 4,
     "sum_p": 2.1555642165984907
    },
    {
     "accuracy": 0.75,
     "correct": 3,
     "gap": -0.11548109529126138,
     "hi": 0.7,
     "lo": 0.6,
     "mean_p": 0.6345189047087386,
     "n": 4,
     "sum_p": 2.5380756188349545
    },
    {
     "accuracy": 0.6666666666666666,
     "correct": 6,
     "gap": 0.0760833781546233,
     "hi": 0.8,
     "lo": 0.7,
     "mean_p": 0.7427500448212899,
     "n": 9,
     "sum_p": 6.684750403391609
    },
    {
     "accuracy": 0.5555555555555556,
     "correct": 10,
     "gap": 0.30610983652377544,
     "hi": 0.9,
     "lo": 0.8,
     "mean_p": 0.861665392079331,
     "n": 18,
     "sum_p": 15.509977057427959
    },
    {
     "accuracy": 0.8461538461538461,
     "correct": 11,
     "gap": 0.06899763158947536,
     "hi": 1.0,
     "lo": 0.9,
     "mean_p": 0.9151514777433215,
     "n": 13,
     "sum_p": 11.89696921066318
    }
   ],
   "unparsed": 0,
   "wall_seconds": 9.315970323999863
  },
  "kf": {
   "accuracy": 0.975,
   "brier": 0.05291543047761599,
   "correct": 39,
   "energy": {
    "cards": [
     "GPU-fp:33a0b4acb8e8"
    ],
    "decisions": 40,
    "joules": 1004.9751100349426,
    "joules_per_decision": 25.124377750873567,
    "mean_watts": 250.86747366653134,
    "net_joules_per_decision": 11.363525499203144,
    "peak_watts": 299.42,
    "per_card": {
     "GPU-fp:33a0b4acb8e8": {
      "joules": 1004.9751100349426,
      "mean_watts": 250.86747366653134,
      "min_watts": 121.03,
      "peak_watts": 299.42,
      "samples": 5,
      "seconds": 4.00600004196167
     }
    },
    "samples": 5,
    "seconds": 4.00600004196167
   },
   "errors": 0,
   "eval_tokens_total": 0,
   "floored_rows": 0,
   "idle": {
    "cards": [
     "GPU-fp:33a0b4acb8e8"
    ],
    "joules": 1237.5835697710513,
    "mean_watts": 137.40241744912183,
    "peak_watts": 237.25,
    "per_card": {
     "GPU-fp:33a0b4acb8e8": {
      "joules": 1237.5835697710513,
      "mean_watts": 137.40241744912183,
      "min_watts": 121.41,
      "peak_watts": 237.25,
      "samples": 10,
      "seconds": 9.006999969482422
     }
    },
    "samples": 10,
    "seconds": 9.006999969482422,
    "what": "the draw of these cards immediately before the arm, with the model resident and no work in flight"
   },
   "letters_floored_total": 0,
   "median_seconds": 0.10945199349953327,
   "n": 40,
   "n_labelled": 40,
   "n_reached": 40,
   "p95_seconds": 0.11135336299957999,
   "prompt_tokens_total": 11036,
   "refused": 0,
   "refused_by_reason": {},
   "reliability": [
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.1,
     "lo": 0.0,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.2,
     "lo": 0.1,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.3,
     "lo": 0.2,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.4,
     "lo": 0.3,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.5,
     "lo": 0.4,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": 1.0,
     "correct": 1,
     "gap": -0.41827784196800855,
     "hi": 0.6,
     "lo": 0.5,
     "mean_p": 0.5817221580319915,
     "n": 1,
     "sum_p": 0.5817221580319915
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.7,
     "lo": 0.6,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": 1.0,
     "correct": 1,
     "gap": -0.21665970028168247,
     "hi": 0.8,
     "lo": 0.7,
     "mean_p": 0.7833402997183175,
     "n": 1,
     "sum_p": 0.7833402997183175
    },
    {
     "accuracy": 0.75,
     "correct": 3,
     "gap": 0.1162090124163645,
     "hi": 0.9,
     "lo": 0.8,
     "mean_p": 0.8662090124163645,
     "n": 4,
     "sum_p": 3.464836049665458
    },
    {
     "accuracy": 1.0,
     "correct": 34,
     "gap": -0.04756930895814948,
     "hi": 1.0,
     "lo": 0.9,
     "mean_p": 0.9524306910418505,
     "n": 34,
     "sum_p": 32.38264349542292
    }
   ],
   "unparsed": 0,
   "wall_seconds": 4.31638923599894
  },
  "kj": {
   "accuracy": 0.7478991596638656,
   "brier": 0.22093638603986238,
   "correct": 89,
   "energy": {
    "cards": [
     "GPU-fp:33a0b4acb8e8"
    ],
    "decisions": 119,
    "joules": 18943.26664948106,
    "joules_per_decision": 159.1871147015215,
    "mean_watts": 295.9053185476013,
    "net_joules_per_decision": 85.81810151408212,
    "peak_watts": 299.82,
    "per_card": {
     "GPU-fp:33a0b4acb8e8": {
      "joules": 18943.26664948106,
      "mean_watts": 295.9053185476013,
      "min_watts": 118.72,
      "peak_watts": 299.82,
      "samples": 65,
      "seconds": 64.01799988746643
     }
    },
    "samples": 65,
    "seconds": 64.01799988746643
   },
   "errors": 0,
   "eval_tokens_total": 0,
   "floored_rows": 0,
   "idle": {
    "cards": [
     "GPU-fp:33a0b4acb8e8"
    ],
    "joules": 1228.5304080617427,
    "mean_watts": 136.3821516550479,
    "peak_watts": 252.12,
    "per_card": {
     "GPU-fp:33a0b4acb8e8": {
      "joules": 1228.5304080617427,
      "mean_watts": 136.3821516550479,
      "min_watts": 119.0,
      "peak_watts": 252.12,
      "samples": 10,
      "seconds": 9.007999897003174
     }
    },
    "samples": 10,
    "seconds": 9.007999897003174,
    "what": "the draw of these cards immediately before the arm, with the model resident and no work in flight"
   },
   "letters_floored_total": 0,
   "median_seconds": 0.49876804199993785,
   "n": 119,
   "n_labelled": 119,
   "n_reached": 119,
   "p95_seconds": 0.9581953390006674,
   "prompt_tokens_total": 137030,
   "refused": 0,
   "refused_by_reason": {},
   "reliability": [
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.1,
     "lo": 0.0,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.2,
     "lo": 0.1,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.3,
     "lo": 0.2,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": 0.0,
     "correct": 0,
     "gap": 0.3666301270074753,
     "hi": 0.4,
     "lo": 0.3,
     "mean_p": 0.3666301270074753,
     "n": 2,
     "sum_p": 0.7332602540149507
    },
    {
     "accuracy": 0.0,
     "correct": 0,
     "gap": 0.44263919571028754,
     "hi": 0.5,
     "lo": 0.4,
     "mean_p": 0.44263919571028754,
     "n": 3,
     "sum_p": 1.3279175871308626
    },
    {
     "accuracy": 0.375,
     "correct": 3,
     "gap": 0.16560721442504545,
     "hi": 0.6,
     "lo": 0.5,
     "mean_p": 0.5406072144250454,
     "n": 8,
     "sum_p": 4.324857715400364
    },
    {
     "accuracy": 0.5,
     "correct": 2,
     "gap": 0.1504808578266511,
     "hi": 0.7,
     "lo": 0.6,
     "mean_p": 0.6504808578266511,
     "n": 4,
     "sum_p": 2.6019234313066044
    },
    {
     "accuracy": 0.18181818181818182,
     "correct": 2,
     "gap": 0.5795918278278367,
     "hi": 0.8,
     "lo": 0.7,
     "mean_p": 0.7614100096460185,
     "n": 11,
     "sum_p": 8.375510106106203
    },
    {
     "accuracy": 0.7,
     "correct": 21,
     "gap": 0.15704946382143115,
     "hi": 0.9,
     "lo": 0.8,
     "mean_p": 0.8570494638214311,
     "n": 30,
     "sum_p": 25.711483914642933
    },
    {
     "accuracy": 1.0,
     "correct": 61,
     "gap": -0.07480583596749046,
     "hi": 1.0,
     "lo": 0.9,
     "mean_p": 0.9251941640325095,
     "n": 61,
     "sum_p": 56.43684400598308
    }
   ],
   "unparsed": 0,
   "wall_seconds": 64.68383760500001
  },
  "kperm": {
   "accuracy": 0.6111111111111112,
   "brier": 0.5401864684760486,
   "correct": 66,
   "energy": {
    "cards": [
     "GPU-fp:33a0b4acb8e8"
    ],
    "decisions": 108,
    "joules": 41724.13047093868,
    "joules_per_decision": 386.3345413975804,
    "mean_watts": 297.9635259411106,
    "net_joules_per_decision": 202.30538335004826,
    "peak_watts": 300.02,
    "per_card": {
     "GPU-fp:33a0b4acb8e8": {
      "joules": 41724.13047093868,
      "mean_watts": 297.9635259411106,
      "min_watts": 124.02,
      "peak_watts": 300.02,
      "samples": 141,
      "seconds": 140.03099989891052
     }
    },
    "samples": 141,
    "seconds": 140.03099989891052
   },
   "errors": 0,
   "eval_tokens_total": 0,
   "floored_rows": 0,
   "idle": {
    "cards": [
     "GPU-fp:33a0b4acb8e8"
    ],
    "joules": 1278.6826817309857,
    "mean_watts": 141.9339223706286,
    "peak_watts": 255.58,
    "per_card": {
     "GPU-fp:33a0b4acb8e8": {
      "joules": 1278.6826817309857,
      "mean_watts": 141.9339223706286,
      "min_watts": 124.44,
      "peak_watts": 255.58,
      "samples": 10,
      "seconds": 9.008999824523926
     }
    },
    "samples": 10,
    "seconds": 9.008999824523926,
    "what": "the draw of these cards immediately before the arm, with the model resident and no work in flight"
   },
   "letters_floored_total": 0,
   "median_seconds": 0.22177773866633288,
   "n": 108,
   "n_labelled": 108,
   "n_reached": 108,
   "p95_seconds": 0.2549782188334575,
   "prompt_tokens_total": 63578,
   "refused": 0,
   "refused_by_reason": {},
   "reliability": [
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.1,
     "lo": 0.0,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.2,
     "lo": 0.1,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": null,
     "correct": 0,
     "gap": null,
     "hi": 0.3,
     "lo": 0.2,
     "mean_p": null,
     "n": 0,
     "sum_p": 0.0
    },
    {
     "accuracy": 0.2,
     "correct": 1,
     "gap": 0.15864685605364082,
     "hi": 0.4,
     "lo": 0.3,
     "mean_p": 0.35864685605364083,
     "n": 5,
     "sum_p": 1.7932342802682042
    },
    {
     "accuracy": 0.23076923076923078,
     "correct": 3,
     "gap": 0.22165952813579248,
     "hi": 0.5,
     "lo": 0.4,
     "mean_p": 0.45242875890502326,
     "n": 13,
     "sum_p": 5.8815738657653025
    },
    {
     "accuracy": 0.6,
     "correct": 6,
     "gap": -0.03067738675275511,
     "hi": 0.6,
     "lo": 0.5,
     "mean_p": 0.5693226132472449,
     "n": 10,
     "sum_p": 5.693226132472448
    },
    {
     "accuracy": 0.25,
     "correct": 3,
     "gap": 0.39860477874718614,
     "hi": 0.7,
     "lo": 0.6,
     "mean_p": 0.6486047787471861,
     "n": 12,
     "sum_p": 7.783257344966234
    },
    {
     "accuracy": 0.5555555555555556,
     "correct": 5,
     "gap": 0.20864407834304366,
     "hi": 0.8,
     "lo": 0.7,
     "mean_p": 0.7641996338985992,
     "n": 9,
     "sum_p": 6.877796705087393
    },
    {
     "accuracy": 0.8125,
     "correct": 26,
     "gap": 0.0453301670260049,
     "hi": 0.9,
     "lo": 0.8,
     "mean_p": 0.8578301670260049,
     "n": 32,
     "sum_p": 27.450565344832157
    },
    {
     "accuracy": 0.8148148148148148,
     "correct": 22,
     "gap": 0.13125102793485321,
     "hi": 1.0,
     "lo": 0.9,
     "mean_p": 0.946065842749668,
     "n": 27,
     "sum_p": 25.543777754241034
    }
   ],
   "unparsed": 0,
   "wall_seconds": 140.10689964000085
  }
 },
 "tag": "counted",
 "tasks": [
  "kc",
  "ka",
  "kd",
  "kf",
  "kj",
  "kperm"
 ],
 "tasks_ended": "2026-09-28T13:19:37Z",
 "tasks_started": "2026-09-28T13:12:44Z",
 "versions": {
  "accelerate": "1.15.0",
  "cuda_devices": 1,
  "cudnn": 92400,
  "device_uuid_fp": "33a0b4acb8e8",
  "dists": [
   "Jinja2==3.1.6",
   "MarkupSafe==3.0.3",
   "PyYAML==6.0.3",
   "Pygments==2.21.0",
   "accelerate==1.15.0",
   "annotated-doc==0.0.5",
   "annotated-types==0.8.0",
   "anyio==4.15.1",
   "certifi==2026.7.22",
   "click==8.5.0",
   "cloudpickle==3.1.2",
   "cuda-bindings==13.4.2",
   "cuda-pathfinder==1.8.2",
   "cuda-toolkit==13.0.3.0",
   "fastapi==0.141.1",
   "filelock==4.0.1",
   "fsspec==2026.9.0",
   "h11==0.16.0",
   "hf-xet==1.6.0",
   "httpcore==1.0.9",
   "httpx==0.28.1",
   "huggingface_hub==1.32.0",
   "idna==3.20",
   "joblib==1.6.0",
   "markdown-it-py==4.2.0",
   "mdurl==0.1.2",
   "mpmath==1.3.0",
   "narwhals==2.26.0",
   "networkx==3.6.1",
   "numpy==2.5.3",
   "nvidia-cublas==13.1.1.3",
   "nvidia-cuda-cupti==13.0.85",
   "nvidia-cuda-nvrtc==13.0.88",
   "nvidia-cuda-runtime==13.0.96",
   "nvidia-cudnn-cu13==9.24.0.43",
   "nvidia-cufft==12.0.0.61",
   "nvidia-cufile==1.15.1.6",
   "nvidia-curand==10.4.0.35",
   "nvidia-cusolver==12.0.4.66",
   "nvidia-cusparse==12.6.3.3",
   "nvidia-cusparselt-cu13==0.8.1",
   "nvidia-nccl-cu13==2.30.7",
   "nvidia-nvjitlink==13.4.92",
   "nvidia-nvshmem-cu13==3.4.5",
   "nvidia-nvtx==13.0.85",
   "packaging==26.3",
   "peft==0.21.0",
   "pillow==12.3.0",
   "pip==25.1.1",
   "pip==26.2.1",
   "psutil==7.2.2",
   "pydantic==2.13.5",
   "pydantic_core==2.46.5",
   "regex==2026.9.10",
   "rich==15.0.0",
   "safetensors==0.8.0",
   "scikit-learn==1.9.1",
   "scipy==1.18.1",
   "sentence-transformers==6.1.0",
   "setuptools==84.0.0",
   "shellingham==1.5.4",
   "starlette==1.7.0",
   "sympy==1.14.0",
   "threadpoolctl==3.7.0",
   "tokenizers==0.23.2",
   "torch==2.14.0",
   "tqdm==4.70.1",
   "transformers==5.17.0",
   "triton==3.8.0",
   "typer==0.27.2",
   "typing-inspection==0.4.4",
   "typing_extensions==4.16.0",
   "uvicorn==0.54.0"
  ],
  "fastapi": "0.141.1",
  "huggingface_hub": "1.32.0",
  "numpy": "2.5.3",
  "peft": "0.21.0",
  "pydantic": "2.13.5",
  "python": "3.14.4",
  "safetensors": "0.8.0",
  "starlette": "1.7.0",
  "tokenizers": "0.23.2",
  "torch": "2.14.0+cu130",
  "torch_cuda": "13.0",
  "transformers": "5.17.0",
  "uvicorn": "0.54.0"
 },
 "warmups_s": [
  0.56,
  0.182,
  0.183
 ],
 "weights": {
  "all_equal_l1_receipts": true,
  "files": 14,
  "residency": {
   "adapter_and_head_bytes": 184601885,
   "bar_mib": 7787.996815824508,
   "dropped_bytes": 908227584,
   "per_prefix_bytes": {
    "model.language_model": 8411510272,
    "model.visual": 667028480,
    "mtp.fc": 26214400,
    "mtp.layers": 214969344,
    "mtp.norm": 5120,
    "mtp.pre_fc_norm_embedding": 5120,
    "mtp.pre_fc_norm_hidden": 5120
   },
   "text_bytes": 8411510272
  }
 }
}