checkpoint-344/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.20873786407767,
  "eval_steps": 500,
  "global_step": 344,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "learning_rate": 2.4848484848484847e-05,
      "loss": 3.4366,
      "step": 41
    },
    {
      "epoch": 1.19,
      "learning_rate": 4.9696969696969694e-05,
      "loss": 1.3573,
      "step": 82
    },
    {
      "epoch": 2.18,
      "learning_rate": 7.454545454545455e-05,
      "loss": 0.3602,
      "step": 123
    },
    {
      "epoch": 3.17,
      "learning_rate": 9.939393939393939e-05,
      "loss": 0.1629,
      "step": 164
    },
    {
      "epoch": 4.16,
      "learning_rate": 9.730276466621713e-05,
      "loss": 0.1326,
      "step": 205
    },
    {
      "epoch": 5.15,
      "learning_rate": 9.453809844908968e-05,
      "loss": 0.1187,
      "step": 246
    },
    {
      "epoch": 6.14,
      "learning_rate": 9.177343223196225e-05,
      "loss": 0.1073,
      "step": 287
    },
    {
      "epoch": 7.13,
      "learning_rate": 8.90087660148348e-05,
      "loss": 0.0997,
      "step": 328
    }
  ],
  "logging_steps": 41,
  "max_steps": 1648,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 8.094040889229312e+16,
  "train_batch_size": 3,
  "trial_name": null,
  "trial_params": null
}