checkpoint-100/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.34873583260680036,
  "eval_steps": 50,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17436791630340018,
      "grad_norm": 0.31028223037719727,
      "learning_rate": 0.0002,
      "loss": 1.7522,
      "mean_token_accuracy": 0.6208396552503109,
      "step": 50
    },
    {
      "epoch": 0.17436791630340018,
      "eval_loss": 1.5662453174591064,
      "eval_mean_token_accuracy": 0.6466238599197537,
      "eval_runtime": 216.8159,
      "eval_samples_per_second": 9.404,
      "eval_steps_per_second": 1.176,
      "step": 50
    },
    {
      "epoch": 0.34873583260680036,
      "grad_norm": 0.20454441010951996,
      "learning_rate": 0.0002,
      "loss": 1.4597,
      "mean_token_accuracy": 0.6637434582412243,
      "step": 100
    },
    {
      "epoch": 0.34873583260680036,
      "eval_loss": 1.4656527042388916,
      "eval_mean_token_accuracy": 0.6612562953257093,
      "eval_runtime": 219.0107,
      "eval_samples_per_second": 9.31,
      "eval_steps_per_second": 1.164,
      "step": 100
    }
  ],
  "logging_steps": 50,
  "max_steps": 858,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.421465319971881e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}