run-14/checkpoint-640/trainer_state.json

{
  "best_metric": 0.6933333333333334,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-14/checkpoint-640",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 640,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.9069746732711792,
      "learning_rate": 0.0001514416935217273,
      "loss": 0.5085,
      "step": 160
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.0,
      "eval_loss": 0.4942961037158966,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.8701,
      "eval_samples_per_second": 681.789,
      "eval_steps_per_second": 21.389,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.130017638206482,
      "learning_rate": 0.00012980716587576626,
      "loss": 0.4801,
      "step": 320
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6862745098039216,
      "eval_f1": 0.2673992673992674,
      "eval_loss": 0.4666464626789093,
      "eval_mcc": 0.18544515371789866,
      "eval_precision": 0.6033057851239669,
      "eval_recall": 0.17176470588235293,
      "eval_runtime": 1.8689,
      "eval_samples_per_second": 682.229,
      "eval_steps_per_second": 21.403,
      "step": 320
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.3810486793518066,
      "learning_rate": 0.00010817263822980523,
      "loss": 0.4677,
      "step": 480
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.687843137254902,
      "eval_f1": 0.45776566757493187,
      "eval_loss": 0.46491068601608276,
      "eval_mcc": 0.25237818375033577,
      "eval_precision": 0.5436893203883495,
      "eval_recall": 0.3952941176470588,
      "eval_runtime": 1.8683,
      "eval_samples_per_second": 682.456,
      "eval_steps_per_second": 21.41,
      "step": 480
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.599716067314148,
      "learning_rate": 8.653811058384418e-05,
      "loss": 0.4606,
      "step": 640
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6933333333333334,
      "eval_f1": 0.3450586264656616,
      "eval_loss": 0.4478222727775574,
      "eval_mcc": 0.22240963104490394,
      "eval_precision": 0.5988372093023255,
      "eval_recall": 0.24235294117647058,
      "eval_runtime": 1.8717,
      "eval_samples_per_second": 681.198,
      "eval_steps_per_second": 21.371,
      "step": 640
    }
  ],
  "logging_steps": 500,
  "max_steps": 1280,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 1167021751680.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7926575748257565,
    "learning_rate": 0.00017307622116768837,
    "num_train_epochs": 8,
    "temperature": 11
  }
}