run-25/checkpoint-244/trainer_state.json

{
  "best_metric": 0.6800391389432485,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-25/checkpoint-244",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 244,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9454018473625183,
      "learning_rate": 3.055872843210697e-05,
      "loss": 0.5488,
      "step": 61
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5,
      "eval_f1": 0.0,
      "eval_loss": 0.5419712662696838,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 28.2925,
      "eval_samples_per_second": 36.123,
      "eval_steps_per_second": 1.131,
      "step": 61
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.9724020957946777,
      "learning_rate": 2.037248562140465e-05,
      "loss": 0.537,
      "step": 122
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5283757338551859,
      "eval_f1": 0.16319444444444445,
      "eval_loss": 0.5242463946342468,
      "eval_precision": 0.7230769230769231,
      "eval_recall": 0.09197651663405088,
      "eval_runtime": 28.1822,
      "eval_samples_per_second": 36.264,
      "eval_steps_per_second": 1.135,
      "step": 122
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.5383687019348145,
      "learning_rate": 1.0186242810702324e-05,
      "loss": 0.5125,
      "step": 183
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6281800391389433,
      "eval_f1": 0.4765840220385675,
      "eval_loss": 0.491844117641449,
      "eval_precision": 0.8046511627906977,
      "eval_recall": 0.3385518590998043,
      "eval_runtime": 28.5655,
      "eval_samples_per_second": 35.777,
      "eval_steps_per_second": 1.12,
      "step": 183
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.672719478607178,
      "learning_rate": 0.0,
      "loss": 0.4914,
      "step": 244
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6800391389432485,
      "eval_f1": 0.6111771700356718,
      "eval_loss": 0.47770681977272034,
      "eval_precision": 0.7787878787878788,
      "eval_recall": 0.50293542074364,
      "eval_runtime": 28.541,
      "eval_samples_per_second": 35.808,
      "eval_steps_per_second": 1.121,
      "step": 244
    }
  ],
  "logging_steps": 500,
  "max_steps": 244,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 942780789120.0,
  "train_batch_size": 51,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7333782587426976,
    "learning_rate": 4.07449712428093e-05,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 51,
    "temperature": 17
  }
}