{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 10160,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.984251968503937,
      "grad_norm": 11363.6455078125,
      "learning_rate": 0.0005993999999999999,
      "loss": 1.4164,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.1411797595750629,
      "eval_loss": 5.487102031707764,
      "eval_runtime": 4.0212,
      "eval_samples_per_second": 53.964,
      "eval_steps_per_second": 1.741,
      "step": 1016
    },
    {
      "epoch": 1.968503937007874,
      "grad_norm": 10689.5205078125,
      "learning_rate": 0.0005345633187772926,
      "loss": 1.259,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.15497758979862383,
      "eval_loss": 5.215289115905762,
      "eval_runtime": 3.6094,
      "eval_samples_per_second": 60.12,
      "eval_steps_per_second": 1.939,
      "step": 2032
    },
    {
      "epoch": 2.952755905511811,
      "grad_norm": 10506.4365234375,
      "learning_rate": 0.000469061135371179,
      "loss": 1.2097,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.16820727407180283,
      "eval_loss": 5.055620193481445,
      "eval_runtime": 3.6003,
      "eval_samples_per_second": 60.272,
      "eval_steps_per_second": 1.944,
      "step": 3048
    },
    {
      "epoch": 3.937007874015748,
      "grad_norm": 10437.658203125,
      "learning_rate": 0.00040355895196506547,
      "loss": 1.1584,
      "step": 4000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.1963981350383724,
      "eval_loss": 4.808844566345215,
      "eval_runtime": 5.4678,
      "eval_samples_per_second": 39.687,
      "eval_steps_per_second": 1.28,
      "step": 4064
    },
    {
      "epoch": 4.921259842519685,
      "grad_norm": 9832.3310546875,
      "learning_rate": 0.000338056768558952,
      "loss": 1.0942,
      "step": 5000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.22212703021995364,
      "eval_loss": 4.599267482757568,
      "eval_runtime": 5.6349,
      "eval_samples_per_second": 38.51,
      "eval_steps_per_second": 1.242,
      "step": 5080
    },
    {
      "epoch": 5.905511811023622,
      "grad_norm": 10148.5263671875,
      "learning_rate": 0.0002725545851528384,
      "loss": 1.0487,
      "step": 6000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.2408578102031798,
      "eval_loss": 4.4708170890808105,
      "eval_runtime": 5.523,
      "eval_samples_per_second": 39.29,
      "eval_steps_per_second": 1.267,
      "step": 6096
    },
    {
      "epoch": 6.889763779527559,
      "grad_norm": 10588.759765625,
      "learning_rate": 0.00020705240174672486,
      "loss": 1.0156,
      "step": 7000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.25204036541704616,
      "eval_loss": 4.3842902183532715,
      "eval_runtime": 3.6379,
      "eval_samples_per_second": 59.649,
      "eval_steps_per_second": 1.924,
      "step": 7112
    },
    {
      "epoch": 7.874015748031496,
      "grad_norm": 11235.5517578125,
      "learning_rate": 0.00014155021834061135,
      "loss": 0.9895,
      "step": 8000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.26269084743928506,
      "eval_loss": 4.315659999847412,
      "eval_runtime": 3.6224,
      "eval_samples_per_second": 59.905,
      "eval_steps_per_second": 1.932,
      "step": 8128
    },
    {
      "epoch": 8.858267716535433,
      "grad_norm": 11518.4150390625,
      "learning_rate": 7.604803493449781e-05,
      "loss": 0.9678,
      "step": 9000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.26984227186234633,
      "eval_loss": 4.2638258934021,
      "eval_runtime": 3.5903,
      "eval_samples_per_second": 60.44,
      "eval_steps_per_second": 1.95,
      "step": 9144
    },
    {
      "epoch": 9.84251968503937,
      "grad_norm": 11941.603515625,
      "learning_rate": 1.054585152838428e-05,
      "loss": 0.9507,
      "step": 10000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.2735848205831162,
      "eval_loss": 4.236743450164795,
      "eval_runtime": 3.5683,
      "eval_samples_per_second": 60.813,
      "eval_steps_per_second": 1.962,
      "step": 10160
    },
    {
      "epoch": 10.0,
      "step": 10160,
      "total_flos": 8.490945871872e+16,
      "train_loss": 1.1083988550141102,
      "train_runtime": 9862.8215,
      "train_samples_per_second": 32.948,
      "train_steps_per_second": 1.03
    }
  ],
  "logging_steps": 1000,
  "max_steps": 10160,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.490945871872e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}