{
  "best_metric": 0.014762421138584614,
  "best_model_checkpoint": "saves/chess/both/checkpoint-1000",
  "epoch": 8.0,
  "eval_steps": 1000,
  "global_step": 4920,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16260162601626016,
      "grad_norm": 1.6060084762973008,
      "learning_rate": 1.0162601626016261e-06,
      "loss": 1.0073,
      "step": 100
    },
    {
      "epoch": 0.3252032520325203,
      "grad_norm": 3.0440875723802594,
      "learning_rate": 2.0325203252032523e-06,
      "loss": 0.0302,
      "step": 200
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 0.9455772942098933,
      "learning_rate": 3.0487804878048782e-06,
      "loss": 0.0229,
      "step": 300
    },
    {
      "epoch": 0.6504065040650406,
      "grad_norm": 0.3831361794753519,
      "learning_rate": 4.0650406504065046e-06,
      "loss": 0.0197,
      "step": 400
    },
    {
      "epoch": 0.8130081300813008,
      "grad_norm": 0.5841201203132319,
      "learning_rate": 4.999959730768458e-06,
      "loss": 0.0183,
      "step": 500
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 0.36973884839665605,
      "learning_rate": 4.992664502959351e-06,
      "loss": 0.0172,
      "step": 600
    },
    {
      "epoch": 1.1382113821138211,
      "grad_norm": 0.5308905808120226,
      "learning_rate": 4.9728272933003704e-06,
      "loss": 0.0149,
      "step": 700
    },
    {
      "epoch": 1.3008130081300813,
      "grad_norm": 0.3199255336490814,
      "learning_rate": 4.940547913829274e-06,
      "loss": 0.015,
      "step": 800
    },
    {
      "epoch": 1.4634146341463414,
      "grad_norm": 0.4781960951795096,
      "learning_rate": 4.89598878006206e-06,
      "loss": 0.0147,
      "step": 900
    },
    {
      "epoch": 1.6260162601626016,
      "grad_norm": 0.5141788877979915,
      "learning_rate": 4.839374093790139e-06,
      "loss": 0.0144,
      "step": 1000
    },
    {
      "epoch": 1.6260162601626016,
      "eval_loss": 0.014762421138584614,
      "eval_runtime": 232.6693,
      "eval_samples_per_second": 150.265,
      "eval_steps_per_second": 0.589,
      "step": 1000
    },
    {
      "epoch": 1.7886178861788617,
      "grad_norm": 0.46618935598211575,
      "learning_rate": 4.770988714996401e-06,
      "loss": 0.0135,
      "step": 1100
    },
    {
      "epoch": 1.951219512195122,
      "grad_norm": 0.5151587169163754,
      "learning_rate": 4.691176728566159e-06,
      "loss": 0.0138,
      "step": 1200
    },
    {
      "epoch": 2.113821138211382,
      "grad_norm": 0.31817021965315584,
      "learning_rate": 4.600339713004673e-06,
      "loss": 0.0109,
      "step": 1300
    },
    {
      "epoch": 2.2764227642276422,
      "grad_norm": 0.455194899760826,
      "learning_rate": 4.498934719872278e-06,
      "loss": 0.011,
      "step": 1400
    },
    {
      "epoch": 2.4390243902439024,
      "grad_norm": 0.3846115904845964,
      "learning_rate": 4.387471974103713e-06,
      "loss": 0.0109,
      "step": 1500
    },
    {
      "epoch": 2.6016260162601625,
      "grad_norm": 0.45607541160743026,
      "learning_rate": 4.266512306782628e-06,
      "loss": 0.0113,
      "step": 1600
    },
    {
      "epoch": 2.7642276422764227,
      "grad_norm": 0.25989554289266725,
      "learning_rate": 4.136664333288392e-06,
      "loss": 0.0111,
      "step": 1700
    },
    {
      "epoch": 2.926829268292683,
      "grad_norm": 0.2511070728030906,
      "learning_rate": 3.998581391013531e-06,
      "loss": 0.0104,
      "step": 1800
    },
    {
      "epoch": 3.089430894308943,
      "grad_norm": 0.24155318360631117,
      "learning_rate": 3.8529582520598395e-06,
      "loss": 0.0082,
      "step": 1900
    },
    {
      "epoch": 3.252032520325203,
      "grad_norm": 0.36145185236007943,
      "learning_rate": 3.7005276274534145e-06,
      "loss": 0.0068,
      "step": 2000
    },
    {
      "epoch": 3.252032520325203,
      "eval_loss": 0.01827162876725197,
      "eval_runtime": 223.5741,
      "eval_samples_per_second": 156.378,
      "eval_steps_per_second": 0.613,
      "step": 2000
    },
    {
      "epoch": 3.4146341463414633,
      "grad_norm": 0.34297060011843655,
      "learning_rate": 3.5420564804678583e-06,
      "loss": 0.0077,
      "step": 2100
    },
    {
      "epoch": 3.5772357723577235,
      "grad_norm": 0.6840105339133411,
      "learning_rate": 3.378342167605362e-06,
      "loss": 0.0073,
      "step": 2200
    },
    {
      "epoch": 3.7398373983739837,
      "grad_norm": 0.30055235437579364,
      "learning_rate": 3.21020842665256e-06,
      "loss": 0.0074,
      "step": 2300
    },
    {
      "epoch": 3.902439024390244,
      "grad_norm": 0.4192737034811711,
      "learning_rate": 3.038501231997454e-06,
      "loss": 0.0069,
      "step": 2400
    },
    {
      "epoch": 4.065040650406504,
      "grad_norm": 0.4192262784149686,
      "learning_rate": 2.8640845380616257e-06,
      "loss": 0.0052,
      "step": 2500
    },
    {
      "epoch": 4.227642276422764,
      "grad_norm": 0.6960471269583507,
      "learning_rate": 2.6878359322649085e-06,
      "loss": 0.0029,
      "step": 2600
    },
    {
      "epoch": 4.390243902439025,
      "grad_norm": 0.3999816795268035,
      "learning_rate": 2.510642219394847e-06,
      "loss": 0.0034,
      "step": 2700
    },
    {
      "epoch": 4.5528455284552845,
      "grad_norm": 0.5318936146765854,
      "learning_rate": 2.3333949595984614e-06,
      "loss": 0.0032,
      "step": 2800
    },
    {
      "epoch": 4.715447154471545,
      "grad_norm": 0.551874865947909,
      "learning_rate": 2.1569859824471445e-06,
      "loss": 0.0033,
      "step": 2900
    },
    {
      "epoch": 4.878048780487805,
      "grad_norm": 0.3248081788920533,
      "learning_rate": 1.9823028996459487e-06,
      "loss": 0.003,
      "step": 3000
    },
    {
      "epoch": 4.878048780487805,
      "eval_loss": 0.02072063274681568,
      "eval_runtime": 223.6378,
      "eval_samples_per_second": 156.333,
      "eval_steps_per_second": 0.613,
      "step": 3000
    },
    {
      "epoch": 5.040650406504065,
      "grad_norm": 0.024739170373672136,
      "learning_rate": 1.8102246389652839e-06,
      "loss": 0.0021,
      "step": 3100
    },
    {
      "epoch": 5.203252032520325,
      "grad_norm": 0.2258552274726826,
      "learning_rate": 1.6416170218663446e-06,
      "loss": 0.0007,
      "step": 3200
    },
    {
      "epoch": 5.365853658536586,
      "grad_norm": 0.3296293455992763,
      "learning_rate": 1.4773284070716504e-06,
      "loss": 0.0008,
      "step": 3300
    },
    {
      "epoch": 5.528455284552845,
      "grad_norm": 0.4671709634217062,
      "learning_rate": 1.3181854220003568e-06,
      "loss": 0.0007,
      "step": 3400
    },
    {
      "epoch": 5.691056910569106,
      "grad_norm": 0.8788212030675407,
      "learning_rate": 1.164988803545826e-06,
      "loss": 0.0007,
      "step": 3500
    },
    {
      "epoch": 5.853658536585366,
      "grad_norm": 0.18192782552523573,
      "learning_rate": 1.0185093691228535e-06,
      "loss": 0.0008,
      "step": 3600
    },
    {
      "epoch": 6.016260162601626,
      "grad_norm": 0.33475268532885594,
      "learning_rate": 8.794841382564212e-07,
      "loss": 0.0006,
      "step": 3700
    },
    {
      "epoch": 6.178861788617886,
      "grad_norm": 0.35558413708310876,
      "learning_rate": 7.486126242264469e-07,
      "loss": 0.0001,
      "step": 3800
    },
    {
      "epoch": 6.341463414634147,
      "grad_norm": 0.6415217440841863,
      "learning_rate": 6.265533144273175e-07,
      "loss": 0.0001,
      "step": 3900
    },
    {
      "epoch": 6.504065040650406,
      "grad_norm": 0.4057619591042224,
      "learning_rate": 5.139203571514673e-07,
      "loss": 0.0001,
      "step": 4000
    },
    {
      "epoch": 6.504065040650406,
      "eval_loss": 0.03588717430830002,
      "eval_runtime": 224.5359,
      "eval_samples_per_second": 155.708,
      "eval_steps_per_second": 0.61,
      "step": 4000
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.015517979645165311,
      "learning_rate": 4.1128047146765936e-07,
      "loss": 0.0001,
      "step": 4100
    },
    {
      "epoch": 6.829268292682927,
      "grad_norm": 0.21375939504530958,
      "learning_rate": 3.191500957420626e-07,
      "loss": 0.0001,
      "step": 4200
    },
    {
      "epoch": 6.991869918699187,
      "grad_norm": 0.5625757372660679,
      "learning_rate": 2.3799278914952957e-07,
      "loss": 0.0001,
      "step": 4300
    },
    {
      "epoch": 7.154471544715447,
      "grad_norm": 0.001893242486182687,
      "learning_rate": 1.682168992494923e-07,
      "loss": 0.0,
      "step": 4400
    },
    {
      "epoch": 7.317073170731708,
      "grad_norm": 0.041593607854158006,
      "learning_rate": 1.1017350736221927e-07,
      "loss": 0.0,
      "step": 4500
    },
    {
      "epoch": 7.479674796747967,
      "grad_norm": 0.0678219109325394,
      "learning_rate": 6.415466208337662e-08,
      "loss": 0.0,
      "step": 4600
    },
    {
      "epoch": 7.642276422764228,
      "grad_norm": 0.0002078547810165964,
      "learning_rate": 3.039190982506823e-08,
      "loss": 0.0,
      "step": 4700
    },
    {
      "epoch": 7.804878048780488,
      "grad_norm": 0.002895716916894551,
      "learning_rate": 9.055129777021664e-09,
      "loss": 0.0,
      "step": 4800
    },
    {
      "epoch": 7.967479674796748,
      "grad_norm": 0.004024341830230839,
      "learning_rate": 2.5167914987633825e-10,
      "loss": 0.0,
      "step": 4900
    },
    {
      "epoch": 8.0,
      "step": 4920,
      "total_flos": 971375200174080.0,
      "train_loss": 0.02723053244252349,
      "train_runtime": 62097.9085,
      "train_samples_per_second": 40.537,
      "train_steps_per_second": 0.079
    }
  ],
  "logging_steps": 100,
  "max_steps": 4920,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 971375200174080.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}