run-10/checkpoint-192/trainer_state.json

{
  "best_metric": 0.7720156555772995,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-10/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": 2.535670757293701,
      "learning_rate": 0.00048723007690763744,
      "loss": 0.5767,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7524461839530333,
      "eval_f1": 0.7924528301886793,
      "eval_loss": 0.4912106394767761,
      "eval_precision": 0.6822033898305084,
      "eval_recall": 0.9452054794520548,
      "eval_runtime": 30.3562,
      "eval_samples_per_second": 33.667,
      "eval_steps_per_second": 1.054,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.952514171600342,
      "learning_rate": 0.0,
      "loss": 0.4909,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7720156555772995,
      "eval_f1": 0.8006843455945253,
      "eval_loss": 0.46037518978118896,
      "eval_precision": 0.7112462006079028,
      "eval_recall": 0.9158512720156555,
      "eval_runtime": 30.68,
      "eval_samples_per_second": 33.312,
      "eval_steps_per_second": 1.043,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 192,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.964566953098255,
    "learning_rate": 0.0009744601538152749,
    "num_train_epochs": 2,
    "temperature": 30
  }
}