run-0/checkpoint-192/trainer_state.json

{
  "best_metric": 0.7896281800391389,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-192",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 192,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.8004393577575684,
      "learning_rate": 0.0007697157586205755,
      "loss": 0.4935,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7436399217221135,
      "eval_f1": 0.7689594356261024,
      "eval_loss": 0.448881596326828,
      "eval_precision": 0.6998394863563403,
      "eval_recall": 0.8532289628180039,
      "eval_runtime": 28.6476,
      "eval_samples_per_second": 35.675,
      "eval_steps_per_second": 1.117,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.9879961013793945,
      "learning_rate": 0.0006841917854405115,
      "loss": 0.4571,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7896281800391389,
      "eval_f1": 0.8057813911472448,
      "eval_loss": 0.42557916045188904,
      "eval_precision": 0.7483221476510067,
      "eval_recall": 0.87279843444227,
      "eval_runtime": 28.772,
      "eval_samples_per_second": 35.521,
      "eval_steps_per_second": 1.112,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 960,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.735211359802956,
    "learning_rate": 0.0008552397318006393,
    "num_train_epochs": 10,
    "temperature": 25
  }
}