checkpoint-1000/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.4390243902439024,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.2154720425605774,
      "eval_runtime": 17.0977,
      "eval_samples_per_second": 64.219,
      "eval_steps_per_second": 4.036,
      "step": 410
    },
    {
      "epoch": 1.2195121951219512,
      "grad_norm": 2.2100486755371094,
      "learning_rate": 1.1869918699186992e-05,
      "loss": 0.4673,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.10782822966575623,
      "eval_runtime": 17.1231,
      "eval_samples_per_second": 64.124,
      "eval_steps_per_second": 4.03,
      "step": 820
    },
    {
      "epoch": 2.4390243902439024,
      "grad_norm": 0.8097894787788391,
      "learning_rate": 3.7398373983739838e-06,
      "loss": 0.1029,
      "step": 1000
    }
  ],
  "logging_steps": 500,
  "max_steps": 1230,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4184868077568000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}