{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.507936507936508,
  "eval_steps": 500,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12698412698412698,
      "grad_norm": 0.023018451407551765,
      "learning_rate": 4.95e-06,
      "loss": 0.0122,
      "on_policy_loss": 0.0122,
      "step": 2
    },
    {
      "epoch": 0.25396825396825395,
      "grad_norm": 0.010841956362128258,
      "learning_rate": 4.85e-06,
      "loss": 0.0088,
      "on_policy_loss": 0.0088,
      "step": 4
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 0.01688988134264946,
      "learning_rate": 4.75e-06,
      "loss": 0.0118,
      "on_policy_loss": 0.0118,
      "step": 6
    },
    {
      "epoch": 0.5079365079365079,
      "grad_norm": 0.014377609826624393,
      "learning_rate": 4.65e-06,
      "loss": 0.0098,
      "on_policy_loss": 0.0098,
      "step": 8
    },
    {
      "epoch": 0.6349206349206349,
      "grad_norm": 0.012471411377191544,
      "learning_rate": 4.5500000000000005e-06,
      "loss": 0.0087,
      "on_policy_loss": 0.0087,
      "step": 10
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 0.028301749378442764,
      "learning_rate": 4.450000000000001e-06,
      "loss": 0.0082,
      "on_policy_loss": 0.0082,
      "step": 12
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 0.017431166023015976,
      "learning_rate": 4.350000000000001e-06,
      "loss": 0.0079,
      "on_policy_loss": 0.0079,
      "step": 14
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.010284779593348503,
      "learning_rate": 4.25e-06,
      "loss": 0.0072,
      "on_policy_loss": 0.0083,
      "step": 16
    },
    {
      "epoch": 1.126984126984127,
      "grad_norm": 0.01129322312772274,
      "learning_rate": 4.15e-06,
      "loss": 0.0086,
      "on_policy_loss": 0.0086,
      "step": 18
    },
    {
      "epoch": 1.253968253968254,
      "grad_norm": 0.01077259425073862,
      "learning_rate": 4.05e-06,
      "loss": 0.0063,
      "on_policy_loss": 0.0063,
      "step": 20
    },
    {
      "epoch": 1.380952380952381,
      "grad_norm": 0.012417323887348175,
      "learning_rate": 3.95e-06,
      "loss": 0.0062,
      "on_policy_loss": 0.0062,
      "step": 22
    },
    {
      "epoch": 1.507936507936508,
      "grad_norm": 0.007292534224689007,
      "learning_rate": 3.85e-06,
      "loss": 0.0059,
      "on_policy_loss": 0.0059,
      "step": 24
    },
    {
      "epoch": 1.6349206349206349,
      "grad_norm": 0.012749790214002132,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 0.0053,
      "on_policy_loss": 0.0053,
      "step": 26
    },
    {
      "epoch": 1.7619047619047619,
      "grad_norm": 0.015549280680716038,
      "learning_rate": 3.65e-06,
      "loss": 0.0047,
      "on_policy_loss": 0.0047,
      "step": 28
    },
    {
      "epoch": 1.8888888888888888,
      "grad_norm": 0.01139769982546568,
      "learning_rate": 3.5500000000000003e-06,
      "loss": 0.0054,
      "on_policy_loss": 0.0054,
      "step": 30
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.03397801145911217,
      "learning_rate": 3.45e-06,
      "loss": 0.0031,
      "on_policy_loss": 0.0035,
      "step": 32
    },
    {
      "epoch": 2.126984126984127,
      "grad_norm": 0.016676733270287514,
      "learning_rate": 3.3500000000000005e-06,
      "loss": 0.0031,
      "on_policy_loss": 0.0031,
      "step": 34
    },
    {
      "epoch": 2.253968253968254,
      "grad_norm": 0.016186246648430824,
      "learning_rate": 3.2500000000000002e-06,
      "loss": 0.0011,
      "on_policy_loss": 0.0011,
      "step": 36
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.026313934475183487,
      "learning_rate": 3.1500000000000003e-06,
      "loss": 0.0003,
      "on_policy_loss": 0.0003,
      "step": 38
    },
    {
      "epoch": 2.507936507936508,
      "grad_norm": 0.04907413199543953,
      "learning_rate": 3.05e-06,
      "loss": -0.0016,
      "on_policy_loss": -0.0016,
      "step": 40
    }
  ],
  "logging_steps": 2,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}