trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.875,
      "eval_f1": 0.8659871972156343,
      "eval_loss": 0.45351532101631165,
      "eval_precision": 0.8830437342948624,
      "eval_runtime": 1.4233,
      "eval_samples_per_second": 1405.223,
      "eval_steps_per_second": 11.242,
      "step": 125
    },
    {
      "epoch": 2.0,
      "learning_rate": 1e-05,
      "loss": 0.6861,
      "step": 250
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9235,
      "eval_f1": 0.9235696204112848,
      "eval_loss": 0.22285881638526917,
      "eval_precision": 0.8962061897780575,
      "eval_runtime": 1.4245,
      "eval_samples_per_second": 1403.999,
      "eval_steps_per_second": 11.232,
      "step": 250
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9295,
      "eval_f1": 0.9296354890231745,
      "eval_loss": 0.18811267614364624,
      "eval_precision": 0.9007997000522071,
      "eval_runtime": 1.4253,
      "eval_samples_per_second": 1403.179,
      "eval_steps_per_second": 11.225,
      "step": 375
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.0,
      "loss": 0.1716,
      "step": 500
    }
  ],
  "logging_steps": 250,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 1440685723392000.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}