{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.5294117647058822,
  "eval_steps": 500,
  "global_step": 45,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.39215686274509803,
      "grad_norm": 0.0274658203125,
      "learning_rate": 0.0001,
      "loss": 0.5995,
      "step": 5
    },
    {
      "epoch": 0.7843137254901961,
      "grad_norm": 0.022705078125,
      "learning_rate": 0.0001,
      "loss": 0.5904,
      "step": 10
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.0189208984375,
      "learning_rate": 0.0001,
      "loss": 0.556,
      "step": 15
    },
    {
      "epoch": 1.5686274509803921,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.0001,
      "loss": 0.5464,
      "step": 20
    },
    {
      "epoch": 1.9607843137254903,
      "grad_norm": 0.0257568359375,
      "learning_rate": 0.0001,
      "loss": 0.5259,
      "step": 25
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.0147705078125,
      "learning_rate": 0.0001,
      "loss": 0.5226,
      "step": 30
    },
    {
      "epoch": 2.7450980392156863,
      "grad_norm": 0.029296875,
      "learning_rate": 0.0001,
      "loss": 0.5098,
      "step": 35
    },
    {
      "epoch": 3.1372549019607843,
      "grad_norm": 0.016357421875,
      "learning_rate": 0.0001,
      "loss": 0.5106,
      "step": 40
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.0252685546875,
      "learning_rate": 0.0001,
      "loss": 0.4622,
      "step": 45
    },
    {
      "epoch": 3.5294117647058822,
      "step": 45,
      "total_flos": 3.685541393109811e+17,
      "train_loss": 0.5359265327453613,
      "train_runtime": 11061.1318,
      "train_samples_per_second": 0.521,
      "train_steps_per_second": 0.004
    }
  ],
  "logging_steps": 5,
  "max_steps": 45,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 180,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.685541393109811e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}