trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 255,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8695652173913043,
      "eval_f1": 0.8728600541101488,
      "eval_loss": 0.6608906984329224,
      "eval_runtime": 0.8181,
      "eval_samples_per_second": 421.7,
      "eval_steps_per_second": 7.334,
      "step": 51
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9188405797101449,
      "eval_f1": 0.9254201727996807,
      "eval_loss": 0.282185822725296,
      "eval_runtime": 0.8009,
      "eval_samples_per_second": 430.79,
      "eval_steps_per_second": 7.492,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9565217391304348,
      "eval_f1": 0.957832432085824,
      "eval_loss": 0.14275817573070526,
      "eval_runtime": 0.7983,
      "eval_samples_per_second": 432.18,
      "eval_steps_per_second": 7.516,
      "step": 153
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9623188405797102,
      "eval_f1": 0.962796387059804,
      "eval_loss": 0.11694291979074478,
      "eval_runtime": 0.7991,
      "eval_samples_per_second": 431.736,
      "eval_steps_per_second": 7.508,
      "step": 204
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9623188405797102,
      "eval_f1": 0.9634447984007641,
      "eval_loss": 0.1356935352087021,
      "eval_runtime": 0.8116,
      "eval_samples_per_second": 425.078,
      "eval_steps_per_second": 7.393,
      "step": 255
    }
  ],
  "logging_steps": 500,
  "max_steps": 255,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 49293495228288.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}