run-1/checkpoint-1532/trainer_state.json

{
  "best_metric": 0.5430528375733855,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-1532",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1532,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.4840396046638489,
      "learning_rate": 3.344893789847612e-05,
      "loss": 0.1172,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5401174168297456,
      "eval_f1": 0.20068027210884357,
      "eval_loss": 0.10136038810014725,
      "eval_precision": 0.7662337662337663,
      "eval_recall": 0.11545988258317025,
      "eval_runtime": 133.3642,
      "eval_samples_per_second": 7.663,
      "eval_steps_per_second": 1.92,
      "step": 766
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.713012456893921,
      "learning_rate": 2.8670518198693823e-05,
      "loss": 0.1055,
      "step": 1532
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5430528375733855,
      "eval_f1": 0.21247892074198985,
      "eval_loss": 0.10160499066114426,
      "eval_precision": 0.7682926829268293,
      "eval_recall": 0.1232876712328767,
      "eval_runtime": 132.4691,
      "eval_samples_per_second": 7.715,
      "eval_steps_per_second": 1.933,
      "step": 1532
    }
  ],
  "logging_steps": 500,
  "max_steps": 6128,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.1218013407365569,
    "learning_rate": 3.822735759825843e-05,
    "num_train_epochs": 8,
    "temperature": 6
  }
}