trainer_state.json

{
  "best_metric": 0.5764515399932861,
  "best_model_checkpoint": "my_awesome_model/checkpoint-510",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 918,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6685082872928176,
      "eval_loss": 0.8615816235542297,
      "eval_runtime": 1.994,
      "eval_samples_per_second": 90.773,
      "eval_steps_per_second": 6.018,
      "step": 102
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8066298342541437,
      "eval_loss": 0.6153411269187927,
      "eval_runtime": 1.8412,
      "eval_samples_per_second": 98.303,
      "eval_steps_per_second": 6.517,
      "step": 204
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8287292817679558,
      "eval_loss": 0.6224671006202698,
      "eval_runtime": 2.4099,
      "eval_samples_per_second": 75.105,
      "eval_steps_per_second": 4.979,
      "step": 306
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8342541436464088,
      "eval_loss": 0.6203207969665527,
      "eval_runtime": 2.2713,
      "eval_samples_per_second": 79.691,
      "eval_steps_per_second": 5.283,
      "step": 408
    },
    {
      "epoch": 4.901960784313726,
      "grad_norm": 2.156015634536743,
      "learning_rate": 1.5098039215686276e-05,
      "loss": 0.5475,
      "step": 500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.850828729281768,
      "eval_loss": 0.5764515399932861,
      "eval_runtime": 2.1191,
      "eval_samples_per_second": 85.413,
      "eval_steps_per_second": 5.663,
      "step": 510
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.861878453038674,
      "eval_loss": 0.5947481393814087,
      "eval_runtime": 2.0161,
      "eval_samples_per_second": 89.776,
      "eval_steps_per_second": 5.952,
      "step": 612
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8674033149171271,
      "eval_loss": 0.6233269572257996,
      "eval_runtime": 2.2366,
      "eval_samples_per_second": 80.928,
      "eval_steps_per_second": 5.365,
      "step": 714
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8729281767955801,
      "eval_loss": 0.6563280820846558,
      "eval_runtime": 2.1333,
      "eval_samples_per_second": 84.846,
      "eval_steps_per_second": 5.625,
      "step": 816
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8784530386740331,
      "eval_loss": 0.7191643714904785,
      "eval_runtime": 2.1691,
      "eval_samples_per_second": 83.444,
      "eval_steps_per_second": 5.532,
      "step": 918
    }
  ],
  "logging_steps": 500,
  "max_steps": 2040,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 1548260124430500.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}