checkpoint-6/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.6666666666666665,
  "eval_steps": 500,
  "global_step": 6,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 1.6168,
      "step": 1
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00016,
      "loss": 1.6153,
      "step": 2
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00012,
      "loss": 1.5975,
      "step": 3
    },
    {
      "epoch": 1.78,
      "learning_rate": 8e-05,
      "loss": 1.5282,
      "step": 4
    },
    {
      "epoch": 2.22,
      "learning_rate": 4e-05,
      "loss": 1.4773,
      "step": 5
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0,
      "loss": 1.4575,
      "step": 6
    }
  ],
  "logging_steps": 1,
  "max_steps": 6,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2352868099522560.0,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}