{
  "best_metric": 0.8277886497064579,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-480",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 480,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.4932310581207275,
      "learning_rate": 0.00022974033046670533,
      "loss": 0.4667,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7651663405088063,
      "eval_f1": 0.7701149425287357,
      "eval_loss": 0.4088786244392395,
      "eval_precision": 0.7542213883677298,
      "eval_recall": 0.786692759295499,
      "eval_runtime": 30.971,
      "eval_samples_per_second": 32.999,
      "eval_steps_per_second": 1.033,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.3421308994293213,
      "learning_rate": 0.00017230524785002898,
      "loss": 0.4108,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8052837573385518,
      "eval_f1": 0.8202348690153567,
      "eval_loss": 0.39227691292762756,
      "eval_precision": 0.761744966442953,
      "eval_recall": 0.8884540117416829,
      "eval_runtime": 30.7242,
      "eval_samples_per_second": 33.264,
      "eval_steps_per_second": 1.042,
      "step": 192
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.60581374168396,
      "learning_rate": 0.00011487016523335267,
      "loss": 0.3914,
      "step": 288
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8199608610567515,
      "eval_f1": 0.8318098720292504,
      "eval_loss": 0.38357415795326233,
      "eval_precision": 0.7804459691252144,
      "eval_recall": 0.8904109589041096,
      "eval_runtime": 32.3837,
      "eval_samples_per_second": 31.559,
      "eval_steps_per_second": 0.988,
      "step": 288
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.155740737915039,
      "learning_rate": 5.743508261667633e-05,
      "loss": 0.3802,
      "step": 384
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8160469667318982,
      "eval_f1": 0.8160469667318982,
      "eval_loss": 0.37758293747901917,
      "eval_precision": 0.8160469667318982,
      "eval_recall": 0.8160469667318982,
      "eval_runtime": 30.8515,
      "eval_samples_per_second": 33.126,
      "eval_steps_per_second": 1.037,
      "step": 384
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.9643956422805786,
      "learning_rate": 0.0,
      "loss": 0.3716,
      "step": 480
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8277886497064579,
      "eval_f1": 0.8367346938775511,
      "eval_loss": 0.37470024824142456,
      "eval_precision": 0.7954144620811288,
      "eval_recall": 0.8825831702544031,
      "eval_runtime": 31.628,
      "eval_samples_per_second": 32.313,
      "eval_steps_per_second": 1.012,
      "step": 480
    }
  ],
  "logging_steps": 500,
  "max_steps": 480,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1178475986400.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6818846524061118,
    "learning_rate": 0.00028717541308338166,
    "num_train_epochs": 5,
    "temperature": 29
  }
}