trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.9097,
      "step": 1
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00017071067811865476,
      "loss": 1.3311,
      "step": 2
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001,
      "loss": 0.8797,
      "step": 3
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.9289321881345254e-05,
      "loss": 1.1031,
      "step": 4
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0,
      "loss": 0.8357,
      "step": 5
    },
    {
      "epoch": 1.0,
      "step": 5,
      "total_flos": 6505574400.0,
      "train_loss": 1.0118353724479676,
      "train_runtime": 29.872,
      "train_samples_per_second": 0.636,
      "train_steps_per_second": 0.167
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 5,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50000,
  "total_flos": 6505574400.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}