run-32/checkpoint-192/trainer_state.json

{
  "best_metric": 0.8072407045009785,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-32/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": 1.7263739109039307,
      "learning_rate": 0.00011867778606566929,
      "loss": 0.5622,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7710371819960861,
      "eval_f1": 0.8006814310051107,
      "eval_loss": 0.4762067198753357,
      "eval_precision": 0.7088989441930619,
      "eval_recall": 0.9197651663405088,
      "eval_runtime": 25.5548,
      "eval_samples_per_second": 39.993,
      "eval_steps_per_second": 1.252,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8613439798355103,
      "learning_rate": 0.00010384306280746064,
      "loss": 0.4724,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8072407045009785,
      "eval_f1": 0.8273444347063978,
      "eval_loss": 0.43931350111961365,
      "eval_precision": 0.7492063492063492,
      "eval_recall": 0.923679060665362,
      "eval_runtime": 25.0105,
      "eval_samples_per_second": 40.863,
      "eval_steps_per_second": 1.279,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 864,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8475925399213161,
    "learning_rate": 0.00013351250932387796,
    "num_train_epochs": 9,
    "temperature": 10
  }
}