trainer_state.json

{
  "best_metric": 7.571461901534349e-05,
  "best_model_checkpoint": "data/models/variant/final/checkpoint-1341",
  "epoch": 3.0,
  "global_step": 1341,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 1.0,
      "eval_f1": 0.0,
      "eval_loss": 0.003437989391386509,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.0426,
      "eval_samples_per_second": 0.959,
      "eval_steps_per_second": 0.959,
      "step": 447
    },
    {
      "epoch": 1.12,
      "learning_rate": 3.17327140606411e-05,
      "loss": 0.2616,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 1.0,
      "eval_f1": 0.0,
      "eval_loss": 0.0002594970865175128,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.23,
      "eval_samples_per_second": 0.813,
      "eval_steps_per_second": 0.813,
      "step": 894
    },
    {
      "epoch": 2.24,
      "learning_rate": 1.2866653382495381e-05,
      "loss": 0.1202,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 1.0,
      "eval_f1": 0.0,
      "eval_loss": 7.571461901534349e-05,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.073,
      "eval_samples_per_second": 0.932,
      "eval_steps_per_second": 0.932,
      "step": 1341
    }
  ],
  "max_steps": 1341,
  "num_train_epochs": 3,
  "total_flos": 1399256782801920.0,
  "trial_name": null,
  "trial_params": null
}