{
  "best_metric": 5.6849164962768555,
  "best_model_checkpoint": "output_sol_3/checkpoint-19350",
  "epoch": 50.0,
  "eval_steps": 500,
  "global_step": 19350,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 9.8e-07,
      "loss": 6.7291,
      "step": 387
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.1754074882356953,
      "eval_loss": 6.441868782043457,
      "eval_runtime": 6.5661,
      "eval_samples_per_second": 6.549,
      "eval_steps_per_second": 0.914,
      "step": 387
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.6e-07,
      "loss": 6.5666,
      "step": 774
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.18468253426993111,
      "eval_loss": 6.273727893829346,
      "eval_runtime": 6.5673,
      "eval_samples_per_second": 6.548,
      "eval_steps_per_second": 0.914,
      "step": 774
    },
    {
      "epoch": 3.0,
      "learning_rate": 9.399999999999999e-07,
      "loss": 6.4479,
      "step": 1161
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.1917524835754393,
      "eval_loss": 6.139554023742676,
      "eval_runtime": 6.5469,
      "eval_samples_per_second": 6.568,
      "eval_steps_per_second": 0.916,
      "step": 1161
    },
    {
      "epoch": 4.0,
      "learning_rate": 9.2e-07,
      "loss": 6.3439,
      "step": 1548
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.1973675236991066,
      "eval_loss": 6.035923480987549,
      "eval_runtime": 6.5514,
      "eval_samples_per_second": 6.563,
      "eval_steps_per_second": 0.916,
      "step": 1548
    },
    {
      "epoch": 5.0,
      "learning_rate": 9e-07,
      "loss": 6.2577,
      "step": 1935
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.20305076269067268,
      "eval_loss": 5.931780815124512,
      "eval_runtime": 6.5535,
      "eval_samples_per_second": 6.561,
      "eval_steps_per_second": 0.916,
      "step": 1935
    },
    {
      "epoch": 6.0,
      "learning_rate": 8.799999999999999e-07,
      "loss": 6.172,
      "step": 2322
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.20827934256291344,
      "eval_loss": 5.861706256866455,
      "eval_runtime": 6.5488,
      "eval_samples_per_second": 6.566,
      "eval_steps_per_second": 0.916,
      "step": 2322
    },
    {
      "epoch": 7.0,
      "learning_rate": 8.599999999999999e-07,
      "loss": 6.0943,
      "step": 2709
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.21153015526608926,
      "eval_loss": 5.821234226226807,
      "eval_runtime": 6.551,
      "eval_samples_per_second": 6.564,
      "eval_steps_per_second": 0.916,
      "step": 2709
    },
    {
      "epoch": 8.0,
      "learning_rate": 8.399999999999999e-07,
      "loss": 6.0318,
      "step": 3096
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.21318966105162654,
      "eval_loss": 5.7982587814331055,
      "eval_runtime": 6.5518,
      "eval_samples_per_second": 6.563,
      "eval_steps_per_second": 0.916,
      "step": 3096
    },
    {
      "epoch": 9.0,
      "learning_rate": 8.199999999999999e-07,
      "loss": 5.9812,
      "step": 3483
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.21512196230875902,
      "eval_loss": 5.781760215759277,
      "eval_runtime": 6.5556,
      "eval_samples_per_second": 6.559,
      "eval_steps_per_second": 0.915,
      "step": 3483
    },
    {
      "epoch": 10.0,
      "learning_rate": 8e-07,
      "loss": 5.9436,
      "step": 3870
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.21634954193093728,
      "eval_loss": 5.767332553863525,
      "eval_runtime": 6.5502,
      "eval_samples_per_second": 6.565,
      "eval_steps_per_second": 0.916,
      "step": 3870
    },
    {
      "epoch": 11.0,
      "learning_rate": 7.799999999999999e-07,
      "loss": 5.9144,
      "step": 4257
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.21714519538975652,
      "eval_loss": 5.75629997253418,
      "eval_runtime": 6.5503,
      "eval_samples_per_second": 6.565,
      "eval_steps_per_second": 0.916,
      "step": 4257
    },
    {
      "epoch": 12.0,
      "learning_rate": 7.599999999999999e-07,
      "loss": 5.8923,
      "step": 4644
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.21846370683579985,
      "eval_loss": 5.745217323303223,
      "eval_runtime": 6.557,
      "eval_samples_per_second": 6.558,
      "eval_steps_per_second": 0.915,
      "step": 4644
    },
    {
      "epoch": 13.0,
      "learning_rate": 7.4e-07,
      "loss": 5.8723,
      "step": 5031
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.2190092977789902,
      "eval_loss": 5.73870325088501,
      "eval_runtime": 6.5631,
      "eval_samples_per_second": 6.552,
      "eval_steps_per_second": 0.914,
      "step": 5031
    },
    {
      "epoch": 14.0,
      "learning_rate": 7.2e-07,
      "loss": 5.8549,
      "step": 5418
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.21930482620655165,
      "eval_loss": 5.7320146560668945,
      "eval_runtime": 6.5518,
      "eval_samples_per_second": 6.563,
      "eval_steps_per_second": 0.916,
      "step": 5418
    },
    {
      "epoch": 15.0,
      "learning_rate": 7e-07,
      "loss": 5.8427,
      "step": 5805
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.219850417149742,
      "eval_loss": 5.725865364074707,
      "eval_runtime": 6.5565,
      "eval_samples_per_second": 6.558,
      "eval_steps_per_second": 0.915,
      "step": 5805
    },
    {
      "epoch": 16.0,
      "learning_rate": 6.800000000000001e-07,
      "loss": 5.8306,
      "step": 6192
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.220282343313101,
      "eval_loss": 5.721892833709717,
      "eval_runtime": 6.5537,
      "eval_samples_per_second": 6.561,
      "eval_steps_per_second": 0.916,
      "step": 6192
    },
    {
      "epoch": 17.0,
      "learning_rate": 6.6e-07,
      "loss": 5.8176,
      "step": 6579
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.22041874104889858,
      "eval_loss": 5.718621253967285,
      "eval_runtime": 6.5536,
      "eval_samples_per_second": 6.561,
      "eval_steps_per_second": 0.916,
      "step": 6579
    },
    {
      "epoch": 18.0,
      "learning_rate": 6.4e-07,
      "loss": 5.8129,
      "step": 6966
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.22062333765259498,
      "eval_loss": 5.7153215408325195,
      "eval_runtime": 6.5502,
      "eval_samples_per_second": 6.565,
      "eval_steps_per_second": 0.916,
      "step": 6966
    },
    {
      "epoch": 19.0,
      "learning_rate": 6.2e-07,
      "loss": 5.8055,
      "step": 7353
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.22112346268385277,
      "eval_loss": 5.7118988037109375,
      "eval_runtime": 6.5485,
      "eval_samples_per_second": 6.566,
      "eval_steps_per_second": 0.916,
      "step": 7353
    },
    {
      "epoch": 20.0,
      "learning_rate": 6e-07,
      "loss": 5.7954,
      "step": 7740
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.2213053263315829,
      "eval_loss": 5.709749698638916,
      "eval_runtime": 6.5556,
      "eval_samples_per_second": 6.559,
      "eval_steps_per_second": 0.915,
      "step": 7740
    },
    {
      "epoch": 21.0,
      "learning_rate": 5.8e-07,
      "loss": 5.7906,
      "step": 8127
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.22185091727477324,
      "eval_loss": 5.70782995223999,
      "eval_runtime": 6.5498,
      "eval_samples_per_second": 6.565,
      "eval_steps_per_second": 0.916,
      "step": 8127
    },
    {
      "epoch": 22.0,
      "learning_rate": 5.6e-07,
      "loss": 5.7834,
      "step": 8514
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.2221237127463684,
      "eval_loss": 5.706095218658447,
      "eval_runtime": 6.556,
      "eval_samples_per_second": 6.559,
      "eval_steps_per_second": 0.915,
      "step": 8514
    },
    {
      "epoch": 23.0,
      "learning_rate": 5.4e-07,
      "loss": 5.7779,
      "step": 8901
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.22235104230603106,
      "eval_loss": 5.704181671142578,
      "eval_runtime": 6.5374,
      "eval_samples_per_second": 6.578,
      "eval_steps_per_second": 0.918,
      "step": 8901
    },
    {
      "epoch": 24.0,
      "learning_rate": 5.2e-07,
      "loss": 5.7718,
      "step": 9288
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.22264657073359248,
      "eval_loss": 5.70188570022583,
      "eval_runtime": 6.5489,
      "eval_samples_per_second": 6.566,
      "eval_steps_per_second": 0.916,
      "step": 9288
    },
    {
      "epoch": 25.0,
      "learning_rate": 5e-07,
      "loss": 5.7664,
      "step": 9675
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.2231466957648503,
      "eval_loss": 5.700209617614746,
      "eval_runtime": 6.5447,
      "eval_samples_per_second": 6.57,
      "eval_steps_per_second": 0.917,
      "step": 9675
    },
    {
      "epoch": 26.0,
      "learning_rate": 4.8e-07,
      "loss": 5.7616,
      "step": 10062
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.22335129236854667,
      "eval_loss": 5.698944091796875,
      "eval_runtime": 6.5389,
      "eval_samples_per_second": 6.576,
      "eval_steps_per_second": 0.918,
      "step": 10062
    },
    {
      "epoch": 27.0,
      "learning_rate": 4.6e-07,
      "loss": 5.7596,
      "step": 10449
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.22341949123644547,
      "eval_loss": 5.697207927703857,
      "eval_runtime": 6.5507,
      "eval_samples_per_second": 6.564,
      "eval_steps_per_second": 0.916,
      "step": 10449
    },
    {
      "epoch": 28.0,
      "learning_rate": 4.3999999999999997e-07,
      "loss": 5.7553,
      "step": 10836
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.2236013548841756,
      "eval_loss": 5.696023464202881,
      "eval_runtime": 6.5473,
      "eval_samples_per_second": 6.568,
      "eval_steps_per_second": 0.916,
      "step": 10836
    },
    {
      "epoch": 29.0,
      "learning_rate": 4.1999999999999995e-07,
      "loss": 5.7525,
      "step": 11223
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.2235331560162768,
      "eval_loss": 5.694847583770752,
      "eval_runtime": 6.543,
      "eval_samples_per_second": 6.572,
      "eval_steps_per_second": 0.917,
      "step": 11223
    },
    {
      "epoch": 30.0,
      "learning_rate": 4e-07,
      "loss": 5.7488,
      "step": 11610
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.22355588897224307,
      "eval_loss": 5.693871974945068,
      "eval_runtime": 6.5401,
      "eval_samples_per_second": 6.575,
      "eval_steps_per_second": 0.917,
      "step": 11610
    },
    {
      "epoch": 31.0,
      "learning_rate": 3.7999999999999996e-07,
      "loss": 5.7445,
      "step": 11997
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.2236695537520744,
      "eval_loss": 5.693030834197998,
      "eval_runtime": 6.5461,
      "eval_samples_per_second": 6.569,
      "eval_steps_per_second": 0.917,
      "step": 11997
    },
    {
      "epoch": 32.0,
      "learning_rate": 3.6e-07,
      "loss": 5.7409,
      "step": 12384
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.22373775261997317,
      "eval_loss": 5.691904544830322,
      "eval_runtime": 6.5492,
      "eval_samples_per_second": 6.566,
      "eval_steps_per_second": 0.916,
      "step": 12384
    },
    {
      "epoch": 33.0,
      "learning_rate": 3.4000000000000003e-07,
      "loss": 5.7389,
      "step": 12771
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.22373775261997317,
      "eval_loss": 5.691229343414307,
      "eval_runtime": 6.5417,
      "eval_samples_per_second": 6.573,
      "eval_steps_per_second": 0.917,
      "step": 12771
    },
    {
      "epoch": 34.0,
      "learning_rate": 3.2e-07,
      "loss": 5.739,
      "step": 13158
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.2238514173998045,
      "eval_loss": 5.690367221832275,
      "eval_runtime": 6.5476,
      "eval_samples_per_second": 6.567,
      "eval_steps_per_second": 0.916,
      "step": 13158
    },
    {
      "epoch": 35.0,
      "learning_rate": 3e-07,
      "loss": 5.7339,
      "step": 13545
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.22387415035577077,
      "eval_loss": 5.689624309539795,
      "eval_runtime": 6.5489,
      "eval_samples_per_second": 6.566,
      "eval_steps_per_second": 0.916,
      "step": 13545
    },
    {
      "epoch": 36.0,
      "learning_rate": 2.8e-07,
      "loss": 5.731,
      "step": 13932
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.22401054809156834,
      "eval_loss": 5.6890740394592285,
      "eval_runtime": 6.5461,
      "eval_samples_per_second": 6.569,
      "eval_steps_per_second": 0.917,
      "step": 13932
    },
    {
      "epoch": 37.0,
      "learning_rate": 2.6e-07,
      "loss": 5.7285,
      "step": 14319
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.22403328104753462,
      "eval_loss": 5.688539505004883,
      "eval_runtime": 6.5471,
      "eval_samples_per_second": 6.568,
      "eval_steps_per_second": 0.916,
      "step": 14319
    },
    {
      "epoch": 38.0,
      "learning_rate": 2.4e-07,
      "loss": 5.724,
      "step": 14706
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.22410147991543342,
      "eval_loss": 5.6880621910095215,
      "eval_runtime": 6.5514,
      "eval_samples_per_second": 6.564,
      "eval_steps_per_second": 0.916,
      "step": 14706
    },
    {
      "epoch": 39.0,
      "learning_rate": 2.1999999999999998e-07,
      "loss": 5.7241,
      "step": 15093
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.2242151446952647,
      "eval_loss": 5.687371253967285,
      "eval_runtime": 6.5505,
      "eval_samples_per_second": 6.564,
      "eval_steps_per_second": 0.916,
      "step": 15093
    },
    {
      "epoch": 40.0,
      "learning_rate": 2e-07,
      "loss": 5.724,
      "step": 15480
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.22444247425492736,
      "eval_loss": 5.687062740325928,
      "eval_runtime": 6.5565,
      "eval_samples_per_second": 6.558,
      "eval_steps_per_second": 0.915,
      "step": 15480
    },
    {
      "epoch": 41.0,
      "learning_rate": 1.8e-07,
      "loss": 5.7199,
      "step": 15867
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.22446520721089364,
      "eval_loss": 5.686655521392822,
      "eval_runtime": 6.5445,
      "eval_samples_per_second": 6.57,
      "eval_steps_per_second": 0.917,
      "step": 15867
    },
    {
      "epoch": 42.0,
      "learning_rate": 1.6e-07,
      "loss": 5.7212,
      "step": 16254
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.2244197412989611,
      "eval_loss": 5.686285495758057,
      "eval_runtime": 6.5455,
      "eval_samples_per_second": 6.569,
      "eval_steps_per_second": 0.917,
      "step": 16254
    },
    {
      "epoch": 43.0,
      "learning_rate": 1.4e-07,
      "loss": 5.7209,
      "step": 16641
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.22451067312282616,
      "eval_loss": 5.685915946960449,
      "eval_runtime": 6.5461,
      "eval_samples_per_second": 6.569,
      "eval_steps_per_second": 0.917,
      "step": 16641
    },
    {
      "epoch": 44.0,
      "learning_rate": 1.2e-07,
      "loss": 5.7169,
      "step": 17028
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.22451067312282616,
      "eval_loss": 5.685729503631592,
      "eval_runtime": 6.5499,
      "eval_samples_per_second": 6.565,
      "eval_steps_per_second": 0.916,
      "step": 17028
    },
    {
      "epoch": 45.0,
      "learning_rate": 1e-07,
      "loss": 5.7185,
      "step": 17415
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.22469253677055628,
      "eval_loss": 5.68557071685791,
      "eval_runtime": 6.5567,
      "eval_samples_per_second": 6.558,
      "eval_steps_per_second": 0.915,
      "step": 17415
    },
    {
      "epoch": 46.0,
      "learning_rate": 8e-08,
      "loss": 5.7165,
      "step": 17802
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.22469253677055628,
      "eval_loss": 5.685311317443848,
      "eval_runtime": 6.5512,
      "eval_samples_per_second": 6.564,
      "eval_steps_per_second": 0.916,
      "step": 17802
    },
    {
      "epoch": 47.0,
      "learning_rate": 6e-08,
      "loss": 5.7202,
      "step": 18189
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.2248062015503876,
      "eval_loss": 5.685103416442871,
      "eval_runtime": 6.5558,
      "eval_samples_per_second": 6.559,
      "eval_steps_per_second": 0.915,
      "step": 18189
    },
    {
      "epoch": 48.0,
      "learning_rate": 4e-08,
      "loss": 5.7183,
      "step": 18576
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.22482893450635386,
      "eval_loss": 5.684987545013428,
      "eval_runtime": 6.5482,
      "eval_samples_per_second": 6.567,
      "eval_steps_per_second": 0.916,
      "step": 18576
    },
    {
      "epoch": 49.0,
      "learning_rate": 2e-08,
      "loss": 5.7179,
      "step": 18963
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.22482893450635386,
      "eval_loss": 5.684926509857178,
      "eval_runtime": 6.5478,
      "eval_samples_per_second": 6.567,
      "eval_steps_per_second": 0.916,
      "step": 18963
    },
    {
      "epoch": 50.0,
      "learning_rate": 0.0,
      "loss": 5.7164,
      "step": 19350
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.22485166746232013,
      "eval_loss": 5.6849164962768555,
      "eval_runtime": 6.5482,
      "eval_samples_per_second": 6.567,
      "eval_steps_per_second": 0.916,
      "step": 19350
    },
    {
      "epoch": 50.0,
      "step": 19350,
      "total_flos": 8.421798445056e+16,
      "train_loss": 5.863803761405847,
      "train_runtime": 11632.7911,
      "train_samples_per_second": 1.663,
      "train_steps_per_second": 1.663
    }
  ],
  "logging_steps": 500,
  "max_steps": 19350,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 8.421798445056e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}