trainer_state.json

{
  "best_metric": 0.012173255905508995,
  "best_model_checkpoint": "models/checkpoint-585",
  "epoch": 2.996168582375479,
  "global_step": 585,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9789031977498615,
      "eval_loss": 0.11250920593738556,
      "eval_runtime": 6.4785,
      "eval_samples_per_second": 172.263,
      "eval_steps_per_second": 21.61,
      "step": 195
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9973135814938732,
      "eval_loss": 0.014149832539260387,
      "eval_runtime": 6.3842,
      "eval_samples_per_second": 174.807,
      "eval_steps_per_second": 21.929,
      "step": 390
    },
    {
      "epoch": 2.56,
      "learning_rate": 7.264957264957266e-06,
      "loss": 0.6233,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9976642888706521,
      "eval_loss": 0.012173255905508995,
      "eval_runtime": 6.4332,
      "eval_samples_per_second": 173.476,
      "eval_steps_per_second": 21.762,
      "step": 585
    },
    {
      "epoch": 3.0,
      "step": 585,
      "total_flos": 1124510524133376.0,
      "train_loss": 0.5359815748328837,
      "train_runtime": 261.4152,
      "train_samples_per_second": 71.828,
      "train_steps_per_second": 2.238
    }
  ],
  "max_steps": 585,
  "num_train_epochs": 3,
  "total_flos": 1124510524133376.0,
  "trial_name": null,
  "trial_params": null
}