{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.857142857142857,
  "eval_steps": 500,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 12.10239091235565,
      "learning_rate": 0.001,
      "loss": 2.2577,
      "step": 1
    },
    {
      "epoch": 0.5714285714285714,
      "eval_loss": 2.2420599460601807,
      "eval_runtime": 0.207,
      "eval_samples_per_second": 53.143,
      "eval_steps_per_second": 4.831,
      "step": 1
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 11.977258060606532,
      "learning_rate": 0.0008535533905932737,
      "loss": 2.2317,
      "step": 2
    },
    {
      "epoch": 1.7142857142857144,
      "eval_loss": 2.237124443054199,
      "eval_runtime": 0.4926,
      "eval_samples_per_second": 22.329,
      "eval_steps_per_second": 2.03,
      "step": 3
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 11.428072306567266,
      "learning_rate": 0.00014644660940672628,
      "loss": 2.8729,
      "step": 4
    },
    {
      "epoch": 2.857142857142857,
      "eval_loss": 1.4730318784713745,
      "eval_runtime": 0.1729,
      "eval_samples_per_second": 63.61,
      "eval_steps_per_second": 5.783,
      "step": 5
    },
    {
      "epoch": 2.857142857142857,
      "step": 5,
      "total_flos": 466196889600.0,
      "train_loss": 2.3683324337005613,
      "train_runtime": 24.8838,
      "train_samples_per_second": 44.005,
      "train_steps_per_second": 0.201
    }
  ],
  "logging_steps": 2,
  "max_steps": 5,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 466196889600.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}