trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 200,
  "global_step": 1434,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.42,
      "eval_accuracy": 0.32553191489361705,
      "eval_loss": 1.8123600482940674,
      "eval_macro_f1": 0.124166903518118,
      "eval_micro_f1": 0.32553191489361705,
      "eval_runtime": 33.706,
      "eval_samples_per_second": 83.665,
      "eval_steps_per_second": 0.682,
      "step": 200
    },
    {
      "epoch": 0.84,
      "eval_accuracy": 0.32163120567375886,
      "eval_loss": 1.7639718055725098,
      "eval_macro_f1": 0.14556223336608437,
      "eval_micro_f1": 0.32163120567375886,
      "eval_runtime": 33.5096,
      "eval_samples_per_second": 84.155,
      "eval_steps_per_second": 0.686,
      "step": 400
    },
    {
      "epoch": 1.05,
      "learning_rate": 3.2566248256624825e-05,
      "loss": 1.8551,
      "step": 500
    },
    {
      "epoch": 1.26,
      "eval_accuracy": 0.3219858156028369,
      "eval_loss": 1.7540267705917358,
      "eval_macro_f1": 0.14782422014060434,
      "eval_micro_f1": 0.3219858156028369,
      "eval_runtime": 34.4226,
      "eval_samples_per_second": 81.923,
      "eval_steps_per_second": 0.668,
      "step": 600
    },
    {
      "epoch": 1.67,
      "eval_accuracy": 0.46879432624113476,
      "eval_loss": 1.695299744606018,
      "eval_macro_f1": 0.21879212132843792,
      "eval_micro_f1": 0.46879432624113476,
      "eval_runtime": 35.6369,
      "eval_samples_per_second": 79.131,
      "eval_steps_per_second": 0.645,
      "step": 800
    },
    {
      "epoch": 2.09,
      "learning_rate": 1.5132496513249652e-05,
      "loss": 1.5754,
      "step": 1000
    },
    {
      "epoch": 2.09,
      "eval_accuracy": 0.5046099290780142,
      "eval_loss": 1.5334745645523071,
      "eval_macro_f1": 0.29034421431874385,
      "eval_micro_f1": 0.5046099290780142,
      "eval_runtime": 34.2142,
      "eval_samples_per_second": 82.422,
      "eval_steps_per_second": 0.672,
      "step": 1000
    },
    {
      "epoch": 2.51,
      "eval_accuracy": 0.5287234042553192,
      "eval_loss": 1.5159926414489746,
      "eval_macro_f1": 0.33578251623147193,
      "eval_micro_f1": 0.5287234042553192,
      "eval_runtime": 34.6936,
      "eval_samples_per_second": 81.283,
      "eval_steps_per_second": 0.663,
      "step": 1200
    },
    {
      "epoch": 2.93,
      "eval_accuracy": 0.5382978723404256,
      "eval_loss": 1.4883990287780762,
      "eval_macro_f1": 0.34546874245733206,
      "eval_micro_f1": 0.5382978723404256,
      "eval_runtime": 35.0801,
      "eval_samples_per_second": 80.387,
      "eval_steps_per_second": 0.656,
      "step": 1400
    },
    {
      "epoch": 3.0,
      "step": 1434,
      "total_flos": 4067515429406100.0,
      "train_loss": 1.5852480240613014,
      "train_runtime": 2084.5163,
      "train_samples_per_second": 21.988,
      "train_steps_per_second": 0.688
    }
  ],
  "logging_steps": 500,
  "max_steps": 1434,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 4067515429406100.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}