run-23/checkpoint-594/trainer_state.json

{
  "best_metric": 0.6237623762376238,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-23/checkpoint-594",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 594,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.666294813156128,
      "learning_rate": 0.00034226276676608983,
      "loss": 0.688,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5900990099009901,
      "eval_f1": 0.5389755011135857,
      "eval_loss": 0.6901505589485168,
      "eval_mcc": 0.18426952240167496,
      "eval_precision": 0.6142131979695431,
      "eval_recall": 0.4801587301587302,
      "eval_runtime": 0.9313,
      "eval_samples_per_second": 542.229,
      "eval_steps_per_second": 17.18,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.4357942342758179,
      "learning_rate": 0.00029336808579950554,
      "loss": 0.6503,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6237623762376238,
      "eval_f1": 0.6895424836601307,
      "eval_loss": 0.641117513179779,
      "eval_mcc": 0.27448731892832257,
      "eval_precision": 0.5861111111111111,
      "eval_recall": 0.8373015873015873,
      "eval_runtime": 0.9311,
      "eval_samples_per_second": 542.381,
      "eval_steps_per_second": 17.184,
      "step": 594
    }
  ],
  "logging_steps": 500,
  "max_steps": 2376,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 1461402531960.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9955244977733414,
    "learning_rate": 0.00039115744773267407,
    "num_train_epochs": 8,
    "temperature": 44
  }
}