{
  "best_metric": 0.821917808219178,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-3830",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 3830,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.53190279006958,
      "learning_rate": 0.00016492442050886904,
      "loss": 0.4371,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7475538160469667,
      "eval_f1": 0.7465618860510805,
      "eval_loss": 0.4075860381126404,
      "eval_precision": 0.7495069033530573,
      "eval_recall": 0.7436399217221135,
      "eval_runtime": 132.2506,
      "eval_samples_per_second": 7.728,
      "eval_steps_per_second": 1.936,
      "step": 766
    },
    {
      "epoch": 2.0,
      "grad_norm": 15.610742568969727,
      "learning_rate": 0.00014430886794526042,
      "loss": 0.4097,
      "step": 1532
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7818003913894325,
      "eval_f1": 0.81053525913339,
      "eval_loss": 0.3935793936252594,
      "eval_precision": 0.7162162162162162,
      "eval_recall": 0.9334637964774951,
      "eval_runtime": 132.8723,
      "eval_samples_per_second": 7.692,
      "eval_steps_per_second": 1.927,
      "step": 1532
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.8801589012145996,
      "learning_rate": 0.00012369331538165178,
      "loss": 0.3882,
      "step": 2298
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8023483365949119,
      "eval_f1": 0.7971887550200805,
      "eval_loss": 0.39150404930114746,
      "eval_precision": 0.8185567010309278,
      "eval_recall": 0.776908023483366,
      "eval_runtime": 132.9694,
      "eval_samples_per_second": 7.686,
      "eval_steps_per_second": 1.925,
      "step": 2298
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.221484422683716,
      "learning_rate": 0.00010307776281804316,
      "loss": 0.382,
      "step": 3064
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8052837573385518,
      "eval_f1": 0.8039408866995074,
      "eval_loss": 0.38070735335350037,
      "eval_precision": 0.8095238095238095,
      "eval_recall": 0.7984344422700587,
      "eval_runtime": 132.6071,
      "eval_samples_per_second": 7.707,
      "eval_steps_per_second": 1.931,
      "step": 3064
    },
    {
      "epoch": 5.0,
      "grad_norm": 23.843862533569336,
      "learning_rate": 8.246221025443452e-05,
      "loss": 0.3666,
      "step": 3830
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.821917808219178,
      "eval_f1": 0.8266666666666668,
      "eval_loss": 0.3845100998878479,
      "eval_precision": 0.8051948051948052,
      "eval_recall": 0.8493150684931506,
      "eval_runtime": 131.7757,
      "eval_samples_per_second": 7.756,
      "eval_steps_per_second": 1.943,
      "step": 3830
    }
  ],
  "logging_steps": 500,
  "max_steps": 6894,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 1178475986400.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6411063430664922,
    "learning_rate": 0.00018553997307247768,
    "num_train_epochs": 9,
    "temperature": 23
  }
}