{
  "best_metric": 0.5742574257425742,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-5346",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 5346,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.7887440323829651,
      "learning_rate": 0.00015586356006012654,
      "loss": 0.3246,
      "step": 594
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5148514851485149,
      "eval_f1": 0.16382252559726962,
      "eval_loss": 0.3221835792064667,
      "eval_precision": 0.5853658536585366,
      "eval_recall": 0.09523809523809523,
      "eval_runtime": 15.4228,
      "eval_samples_per_second": 32.744,
      "eval_steps_per_second": 1.037,
      "step": 594
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6338050961494446,
      "learning_rate": 0.00013854538672011249,
      "loss": 0.3178,
      "step": 1188
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5168316831683168,
      "eval_f1": 0.17006802721088435,
      "eval_loss": 0.31132203340530396,
      "eval_precision": 0.5952380952380952,
      "eval_recall": 0.0992063492063492,
      "eval_runtime": 15.3986,
      "eval_samples_per_second": 32.795,
      "eval_steps_per_second": 1.039,
      "step": 1188
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.1013684272766113,
      "learning_rate": 0.0001212272133800984,
      "loss": 0.3119,
      "step": 1782
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5207920792079208,
      "eval_f1": 0.1986754966887417,
      "eval_loss": 0.3059851825237274,
      "eval_precision": 0.6,
      "eval_recall": 0.11904761904761904,
      "eval_runtime": 15.2233,
      "eval_samples_per_second": 33.173,
      "eval_steps_per_second": 1.051,
      "step": 1782
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.1249533891677856,
      "learning_rate": 0.00010390904004008435,
      "loss": 0.3068,
      "step": 2376
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5287128712871287,
      "eval_f1": 0.1793103448275862,
      "eval_loss": 0.3041296601295471,
      "eval_precision": 0.6842105263157895,
      "eval_recall": 0.10317460317460317,
      "eval_runtime": 15.2189,
      "eval_samples_per_second": 33.183,
      "eval_steps_per_second": 1.051,
      "step": 2376
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.003902554512024,
      "learning_rate": 8.65908667000703e-05,
      "loss": 0.3031,
      "step": 2970
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5465346534653466,
      "eval_f1": 0.23411371237458192,
      "eval_loss": 0.3002361059188843,
      "eval_precision": 0.7446808510638298,
      "eval_recall": 0.1388888888888889,
      "eval_runtime": 15.2406,
      "eval_samples_per_second": 33.135,
      "eval_steps_per_second": 1.05,
      "step": 2970
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.5786848068237305,
      "learning_rate": 6.927269336005624e-05,
      "loss": 0.3004,
      "step": 3564
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5683168316831683,
      "eval_f1": 0.30128205128205127,
      "eval_loss": 0.2976093292236328,
      "eval_precision": 0.7833333333333333,
      "eval_recall": 0.1865079365079365,
      "eval_runtime": 15.1978,
      "eval_samples_per_second": 33.228,
      "eval_steps_per_second": 1.053,
      "step": 3564
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.880638599395752,
      "learning_rate": 5.1954520020042175e-05,
      "loss": 0.2983,
      "step": 4158
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5584158415841585,
      "eval_f1": 0.29652996845425866,
      "eval_loss": 0.29786166548728943,
      "eval_precision": 0.7230769230769231,
      "eval_recall": 0.1865079365079365,
      "eval_runtime": 15.1508,
      "eval_samples_per_second": 33.332,
      "eval_steps_per_second": 1.056,
      "step": 4158
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.8452739119529724,
      "learning_rate": 3.463634668002812e-05,
      "loss": 0.296,
      "step": 4752
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5722772277227722,
      "eval_f1": 0.3076923076923077,
      "eval_loss": 0.29773908853530884,
      "eval_precision": 0.8,
      "eval_recall": 0.19047619047619047,
      "eval_runtime": 15.0116,
      "eval_samples_per_second": 33.641,
      "eval_steps_per_second": 1.066,
      "step": 4752
    },
    {
      "epoch": 9.0,
      "grad_norm": 1.1398597955703735,
      "learning_rate": 1.731817334001406e-05,
      "loss": 0.2955,
      "step": 5346
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5742574257425742,
      "eval_f1": 0.3260188087774295,
      "eval_loss": 0.2973213791847229,
      "eval_precision": 0.7761194029850746,
      "eval_recall": 0.20634920634920634,
      "eval_runtime": 15.1503,
      "eval_samples_per_second": 33.333,
      "eval_steps_per_second": 1.056,
      "step": 5346
    }
  ],
  "logging_steps": 500,
  "max_steps": 5940,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 7000589548260.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.39860510957002215,
    "learning_rate": 0.0001731817334001406,
    "num_train_epochs": 10,
    "per_device_train_batch_size": 16,
    "temperature": 4
  }
}