{
  "best_metric": 0.969468891620636,
  "best_model_checkpoint": "chessgpt2-medium-m/checkpoint-20000",
  "epoch": 2.559508574353724,
  "eval_steps": 1000,
  "global_step": 20000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1279754287176862,
      "grad_norm": 1.6428743600845337,
      "learning_rate": 4.9959725443266765e-05,
      "loss": 2.4607,
      "step": 1000
    },
    {
      "epoch": 0.1279754287176862,
      "eval_loss": 1.8105345964431763,
      "eval_runtime": 376.5583,
      "eval_samples_per_second": 73.781,
      "eval_steps_per_second": 9.223,
      "step": 1000
    },
    {
      "epoch": 0.2559508574353724,
      "grad_norm": 1.0423442125320435,
      "learning_rate": 4.952992634500169e-05,
      "loss": 1.6726,
      "step": 2000
    },
    {
      "epoch": 0.2559508574353724,
      "eval_loss": 1.4938699007034302,
      "eval_runtime": 376.7393,
      "eval_samples_per_second": 73.746,
      "eval_steps_per_second": 9.219,
      "step": 2000
    },
    {
      "epoch": 0.3839262861530586,
      "grad_norm": 1.234433889389038,
      "learning_rate": 4.863737385256343e-05,
      "loss": 1.4686,
      "step": 3000
    },
    {
      "epoch": 0.3839262861530586,
      "eval_loss": 1.3617429733276367,
      "eval_runtime": 379.1693,
      "eval_samples_per_second": 73.273,
      "eval_steps_per_second": 9.159,
      "step": 3000
    },
    {
      "epoch": 0.5119017148707448,
      "grad_norm": 1.1198837757110596,
      "learning_rate": 4.729890583531792e-05,
      "loss": 1.3601,
      "step": 4000
    },
    {
      "epoch": 0.5119017148707448,
      "eval_loss": 1.2697532176971436,
      "eval_runtime": 376.7562,
      "eval_samples_per_second": 73.743,
      "eval_steps_per_second": 9.218,
      "step": 4000
    },
    {
      "epoch": 0.6398771435884311,
      "grad_norm": 0.9552892446517944,
      "learning_rate": 4.5539772292358576e-05,
      "loss": 1.2875,
      "step": 5000
    },
    {
      "epoch": 0.6398771435884311,
      "eval_loss": 1.2163749933242798,
      "eval_runtime": 378.6404,
      "eval_samples_per_second": 73.376,
      "eval_steps_per_second": 9.172,
      "step": 5000
    },
    {
      "epoch": 0.7678525723061173,
      "grad_norm": 1.001042127609253,
      "learning_rate": 4.3393159015047314e-05,
      "loss": 1.2366,
      "step": 6000
    },
    {
      "epoch": 0.7678525723061173,
      "eval_loss": 1.1716755628585815,
      "eval_runtime": 376.0934,
      "eval_samples_per_second": 73.873,
      "eval_steps_per_second": 9.234,
      "step": 6000
    },
    {
      "epoch": 0.8958280010238034,
      "grad_norm": 1.0304712057113647,
      "learning_rate": 4.089956154202057e-05,
      "loss": 1.1992,
      "step": 7000
    },
    {
      "epoch": 0.8958280010238034,
      "eval_loss": 1.1404258012771606,
      "eval_runtime": 377.955,
      "eval_samples_per_second": 73.509,
      "eval_steps_per_second": 9.189,
      "step": 7000
    },
    {
      "epoch": 1.0238034297414895,
      "grad_norm": 0.7832671403884888,
      "learning_rate": 3.810602121690553e-05,
      "loss": 1.1646,
      "step": 8000
    },
    {
      "epoch": 1.0238034297414895,
      "eval_loss": 1.1096745729446411,
      "eval_runtime": 377.602,
      "eval_samples_per_second": 73.577,
      "eval_steps_per_second": 9.198,
      "step": 8000
    },
    {
      "epoch": 1.151778858459176,
      "grad_norm": 0.7852849364280701,
      "learning_rate": 3.5065237760403566e-05,
      "loss": 1.126,
      "step": 9000
    },
    {
      "epoch": 1.151778858459176,
      "eval_loss": 1.0933990478515625,
      "eval_runtime": 378.2151,
      "eval_samples_per_second": 73.458,
      "eval_steps_per_second": 9.183,
      "step": 9000
    },
    {
      "epoch": 1.279754287176862,
      "grad_norm": 0.7889285087585449,
      "learning_rate": 3.183457509793587e-05,
      "loss": 1.1085,
      "step": 10000
    },
    {
      "epoch": 1.279754287176862,
      "eval_loss": 1.0705418586730957,
      "eval_runtime": 376.2258,
      "eval_samples_per_second": 73.847,
      "eval_steps_per_second": 9.231,
      "step": 10000
    },
    {
      "epoch": 1.4077297158945483,
      "grad_norm": 0.8414849638938904,
      "learning_rate": 2.8478382244815133e-05,
      "loss": 1.0875,
      "step": 11000
    },
    {
      "epoch": 1.4077297158945483,
      "eval_loss": 1.0541125535964966,
      "eval_runtime": 378.6403,
      "eval_samples_per_second": 73.376,
      "eval_steps_per_second": 9.172,
      "step": 11000
    },
    {
      "epoch": 1.5357051446122343,
      "grad_norm": 0.8442362546920776,
      "learning_rate": 2.5053264766232426e-05,
      "loss": 1.0733,
      "step": 12000
    },
    {
      "epoch": 1.5357051446122343,
      "eval_loss": 1.0388243198394775,
      "eval_runtime": 376.572,
      "eval_samples_per_second": 73.779,
      "eval_steps_per_second": 9.223,
      "step": 12000
    },
    {
      "epoch": 1.6636805733299207,
      "grad_norm": 0.9030175805091858,
      "learning_rate": 2.1627142455795886e-05,
      "loss": 1.059,
      "step": 13000
    },
    {
      "epoch": 1.6636805733299207,
      "eval_loss": 1.0245987176895142,
      "eval_runtime": 378.3137,
      "eval_samples_per_second": 73.439,
      "eval_steps_per_second": 9.18,
      "step": 13000
    },
    {
      "epoch": 1.7916560020476069,
      "grad_norm": 0.8050616979598999,
      "learning_rate": 1.8264648598812123e-05,
      "loss": 1.0451,
      "step": 14000
    },
    {
      "epoch": 1.7916560020476069,
      "eval_loss": 1.0108946561813354,
      "eval_runtime": 376.6418,
      "eval_samples_per_second": 73.765,
      "eval_steps_per_second": 9.221,
      "step": 14000
    },
    {
      "epoch": 1.919631430765293,
      "grad_norm": 0.7604677677154541,
      "learning_rate": 1.503551911234875e-05,
      "loss": 1.0327,
      "step": 15000
    },
    {
      "epoch": 1.919631430765293,
      "eval_loss": 1.0017954111099243,
      "eval_runtime": 378.423,
      "eval_samples_per_second": 73.418,
      "eval_steps_per_second": 9.178,
      "step": 15000
    },
    {
      "epoch": 2.047606859482979,
      "grad_norm": 0.8110759258270264,
      "learning_rate": 1.1987749064346765e-05,
      "loss": 1.0118,
      "step": 16000
    },
    {
      "epoch": 2.047606859482979,
      "eval_loss": 0.991495668888092,
      "eval_runtime": 377.8976,
      "eval_samples_per_second": 73.52,
      "eval_steps_per_second": 9.19,
      "step": 16000
    },
    {
      "epoch": 2.1755822882006655,
      "grad_norm": 0.8434863090515137,
      "learning_rate": 9.185453185391116e-06,
      "loss": 0.9862,
      "step": 17000
    },
    {
      "epoch": 2.1755822882006655,
      "eval_loss": 0.9861236810684204,
      "eval_runtime": 378.3771,
      "eval_samples_per_second": 73.427,
      "eval_steps_per_second": 9.179,
      "step": 17000
    },
    {
      "epoch": 2.303557716918352,
      "grad_norm": 0.8797647356987,
      "learning_rate": 6.681496368397716e-06,
      "loss": 0.9806,
      "step": 18000
    },
    {
      "epoch": 2.303557716918352,
      "eval_loss": 0.9783245325088501,
      "eval_runtime": 376.5913,
      "eval_samples_per_second": 73.775,
      "eval_steps_per_second": 9.222,
      "step": 18000
    },
    {
      "epoch": 2.431533145636038,
      "grad_norm": 0.8292227983474731,
      "learning_rate": 4.5231153844693594e-06,
      "loss": 0.9757,
      "step": 19000
    },
    {
      "epoch": 2.431533145636038,
      "eval_loss": 0.9735883474349976,
      "eval_runtime": 372.1389,
      "eval_samples_per_second": 74.658,
      "eval_steps_per_second": 9.333,
      "step": 19000
    },
    {
      "epoch": 2.559508574353724,
      "grad_norm": 0.8722350597381592,
      "learning_rate": 2.75102776826896e-06,
      "loss": 0.9713,
      "step": 20000
    },
    {
      "epoch": 2.559508574353724,
      "eval_loss": 0.969468891620636,
      "eval_runtime": 372.7114,
      "eval_samples_per_second": 74.543,
      "eval_steps_per_second": 9.318,
      "step": 20000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 23442,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7.87023340512215e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}