run-11/checkpoint-192/trainer_state.json

{
  "best_metric": 0.8023483365949119,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-11/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": 8.867341995239258,
      "learning_rate": 0.00047267095070335774,
      "loss": 0.5752,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7514677103718199,
      "eval_f1": 0.765249537892791,
      "eval_loss": 0.501445472240448,
      "eval_precision": 0.7250437828371279,
      "eval_recall": 0.8101761252446184,
      "eval_runtime": 30.0696,
      "eval_samples_per_second": 33.988,
      "eval_steps_per_second": 1.064,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.714812755584717,
      "learning_rate": 0.0,
      "loss": 0.4797,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8023483365949119,
      "eval_f1": 0.8246527777777778,
      "eval_loss": 0.4407924711704254,
      "eval_precision": 0.7410296411856474,
      "eval_recall": 0.9295499021526419,
      "eval_runtime": 29.5981,
      "eval_samples_per_second": 34.529,
      "eval_steps_per_second": 1.081,
      "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": 0.9981740599737018,
    "learning_rate": 0.0009453419014067155,
    "num_train_epochs": 2,
    "temperature": 30
  }
}