run-19/checkpoint-192/trainer_state.json

{
  "best_metric": 0.812133072407045,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-19/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": 3.265742301940918,
      "learning_rate": 0.00016552182978108178,
      "loss": 0.5708,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7592954990215264,
      "eval_f1": 0.7980295566502462,
      "eval_loss": 0.4895075857639313,
      "eval_precision": 0.6874115983026874,
      "eval_recall": 0.9510763209393346,
      "eval_runtime": 24.8356,
      "eval_samples_per_second": 41.151,
      "eval_steps_per_second": 1.288,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.506702899932861,
      "learning_rate": 0.00014713051536096157,
      "loss": 0.4657,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.812133072407045,
      "eval_f1": 0.8318739054290717,
      "eval_loss": 0.42986905574798584,
      "eval_precision": 0.7527733755942948,
      "eval_recall": 0.9295499021526419,
      "eval_runtime": 25.1401,
      "eval_samples_per_second": 40.652,
      "eval_steps_per_second": 1.273,
      "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": 0.8822091648395545,
    "learning_rate": 0.00018391314420120196,
    "num_train_epochs": 10,
    "temperature": 8
  }
}