{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 99.87154784842646,
  "eval_steps": 3110,
  "global_step": 31100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 9.987154784842646,
      "grad_norm": 3.000732898712158,
      "learning_rate": 9.00096463022508e-06,
      "loss": 2.3923,
      "step": 3110
    },
    {
      "epoch": 9.987154784842646,
      "eval_accuracy": 0.0765299914831488,
      "eval_loss": 3.097198724746704,
      "eval_runtime": 31.7938,
      "eval_samples_per_second": 258.51,
      "eval_steps_per_second": 12.927,
      "step": 3110
    },
    {
      "epoch": 19.974309569685293,
      "grad_norm": 7.0140557289123535,
      "learning_rate": 8.00128617363344e-06,
      "loss": 2.1272,
      "step": 6220
    },
    {
      "epoch": 19.974309569685293,
      "eval_accuracy": 0.07817252707141988,
      "eval_loss": 3.585836410522461,
      "eval_runtime": 31.7391,
      "eval_samples_per_second": 258.955,
      "eval_steps_per_second": 12.949,
      "step": 6220
    },
    {
      "epoch": 29.961464354527937,
      "grad_norm": 4.256643295288086,
      "learning_rate": 7.001607717041802e-06,
      "loss": 1.9949,
      "step": 9330
    },
    {
      "epoch": 29.961464354527937,
      "eval_accuracy": 0.07851725676278541,
      "eval_loss": 3.6032633781433105,
      "eval_runtime": 31.708,
      "eval_samples_per_second": 259.209,
      "eval_steps_per_second": 12.962,
      "step": 9330
    },
    {
      "epoch": 39.948619139370585,
      "grad_norm": 3.2846930027008057,
      "learning_rate": 6.0019292604501615e-06,
      "loss": 1.957,
      "step": 12440
    },
    {
      "epoch": 39.948619139370585,
      "eval_accuracy": 0.07692541671736221,
      "eval_loss": 3.7207980155944824,
      "eval_runtime": 31.5878,
      "eval_samples_per_second": 260.195,
      "eval_steps_per_second": 13.011,
      "step": 12440
    },
    {
      "epoch": 49.93577392421323,
      "grad_norm": 1.5283381938934326,
      "learning_rate": 5.002572347266882e-06,
      "loss": 1.9313,
      "step": 15550
    },
    {
      "epoch": 49.93577392421323,
      "eval_accuracy": 0.07587297724784037,
      "eval_loss": 3.817392110824585,
      "eval_runtime": 31.8048,
      "eval_samples_per_second": 258.421,
      "eval_steps_per_second": 12.923,
      "step": 15550
    },
    {
      "epoch": 59.922928709055874,
      "grad_norm": 1.3086577653884888,
      "learning_rate": 4.002893890675241e-06,
      "loss": 1.9255,
      "step": 18660
    },
    {
      "epoch": 59.922928709055874,
      "eval_accuracy": 0.07484689946059943,
      "eval_loss": 3.914490222930908,
      "eval_runtime": 31.7322,
      "eval_samples_per_second": 259.011,
      "eval_steps_per_second": 12.952,
      "step": 18660
    },
    {
      "epoch": 69.91008349389853,
      "grad_norm": 0.9638305902481079,
      "learning_rate": 3.0035369774919613e-06,
      "loss": 1.9179,
      "step": 21770
    },
    {
      "epoch": 69.91008349389853,
      "eval_accuracy": 0.07458328263779049,
      "eval_loss": 4.036659240722656,
      "eval_runtime": 31.8221,
      "eval_samples_per_second": 258.279,
      "eval_steps_per_second": 12.916,
      "step": 21770
    },
    {
      "epoch": 79.89723827874117,
      "grad_norm": 1.957848072052002,
      "learning_rate": 2.0041800643086816e-06,
      "loss": 1.9133,
      "step": 24880
    },
    {
      "epoch": 79.89723827874117,
      "eval_accuracy": 0.0740205621121791,
      "eval_loss": 4.068979740142822,
      "eval_runtime": 31.7024,
      "eval_samples_per_second": 259.255,
      "eval_steps_per_second": 12.964,
      "step": 24880
    },
    {
      "epoch": 89.88439306358381,
      "grad_norm": 0.8189646005630493,
      "learning_rate": 1.0045016077170419e-06,
      "loss": 1.9102,
      "step": 27990
    },
    {
      "epoch": 89.88439306358381,
      "eval_accuracy": 0.07374511632937232,
      "eval_loss": 4.122669219970703,
      "eval_runtime": 31.652,
      "eval_samples_per_second": 259.667,
      "eval_steps_per_second": 12.985,
      "step": 27990
    },
    {
      "epoch": 99.87154784842646,
      "grad_norm": 0.5666019320487976,
      "learning_rate": 5.144694533762058e-09,
      "loss": 1.9084,
      "step": 31100
    },
    {
      "epoch": 99.87154784842646,
      "eval_accuracy": 0.07336658960944153,
      "eval_loss": 4.205153942108154,
      "eval_runtime": 31.8056,
      "eval_samples_per_second": 258.414,
      "eval_steps_per_second": 12.922,
      "step": 31100
    },
    {
      "epoch": 99.87154784842646,
      "step": 31100,
      "total_flos": 2.458812304495296e+18,
      "train_loss": 1.9977928117463826,
      "train_runtime": 55190.6147,
      "train_samples_per_second": 112.8,
      "train_steps_per_second": 0.564
    }
  ],
  "logging_steps": 3110,
  "max_steps": 31100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "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": 2.458812304495296e+18,
  "train_batch_size": 40,
  "trial_name": null,
  "trial_params": null
}