run-10/checkpoint-320/trainer_state.json

{
  "best_metric": 0.676078431372549,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-10/checkpoint-320",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 320,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.012742280960083,
      "learning_rate": 0.0007591954920690624,
      "loss": 0.5869,
      "step": 160
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.0,
      "eval_loss": 0.5785399675369263,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.8697,
      "eval_samples_per_second": 681.924,
      "eval_steps_per_second": 21.394,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.057482957839966,
      "learning_rate": 0.0006642960555604296,
      "loss": 0.5715,
      "step": 320
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.676078431372549,
      "eval_f1": 0.4411366711772666,
      "eval_loss": 0.5550761222839355,
      "eval_mcc": 0.22526645932553852,
      "eval_precision": 0.5191082802547771,
      "eval_recall": 0.3835294117647059,
      "eval_runtime": 1.8672,
      "eval_samples_per_second": 682.827,
      "eval_steps_per_second": 21.422,
      "step": 320
    }
  ],
  "logging_steps": 500,
  "max_steps": 1440,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 583510875840.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9116955099903541,
    "learning_rate": 0.0008540949285776952,
    "num_train_epochs": 9,
    "temperature": 21
  }
}