{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.8235294117647056,
  "eval_steps": 500,
  "global_step": 12,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.24,
      "grad_norm": 7.997311565277474,
      "learning_rate": 1e-05,
      "loss": 0.7301,
      "step": 1
    },
    {
      "epoch": 0.94,
      "eval_loss": 0.7239956259727478,
      "eval_runtime": 55.4169,
      "eval_samples_per_second": 8.463,
      "eval_steps_per_second": 0.541,
      "step": 4
    },
    {
      "epoch": 1.18,
      "grad_norm": 12.593935110474865,
      "learning_rate": 1.5877852522924733e-05,
      "loss": 0.8143,
      "step": 5
    },
    {
      "epoch": 1.88,
      "eval_loss": 0.5195969939231873,
      "eval_runtime": 55.9511,
      "eval_samples_per_second": 8.382,
      "eval_steps_per_second": 0.536,
      "step": 8
    },
    {
      "epoch": 2.35,
      "grad_norm": 3.1734091459606377,
      "learning_rate": 1.9098300562505266e-06,
      "loss": 0.5734,
      "step": 10
    },
    {
      "epoch": 2.82,
      "eval_loss": 0.48763999342918396,
      "eval_runtime": 55.7091,
      "eval_samples_per_second": 8.419,
      "eval_steps_per_second": 0.539,
      "step": 12
    },
    {
      "epoch": 2.82,
      "step": 12,
      "total_flos": 2460210954240.0,
      "train_loss": 0.6499238312244415,
      "train_runtime": 440.7059,
      "train_samples_per_second": 1.749,
      "train_steps_per_second": 0.027
    }
  ],
  "logging_steps": 5,
  "max_steps": 12,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2460210954240.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}