{
  "best_metric": 0.8160469667318982,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-9/checkpoint-405",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 486,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9513042569160461,
      "learning_rate": 0.0005954099159667608,
      "loss": 0.499,
      "step": 81
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6829745596868885,
      "eval_f1": 0.6067961165048544,
      "eval_loss": 0.4608267843723297,
      "eval_precision": 0.7987220447284346,
      "eval_recall": 0.4892367906066536,
      "eval_runtime": 28.3141,
      "eval_samples_per_second": 36.095,
      "eval_steps_per_second": 1.13,
      "step": 81
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.7321057319641113,
      "learning_rate": 0.0004961749299723006,
      "loss": 0.4578,
      "step": 162
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7827788649706457,
      "eval_f1": 0.798913043478261,
      "eval_loss": 0.42703160643577576,
      "eval_precision": 0.7436762225969646,
      "eval_recall": 0.863013698630137,
      "eval_runtime": 28.5611,
      "eval_samples_per_second": 35.783,
      "eval_steps_per_second": 1.12,
      "step": 162
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.0545783042907715,
      "learning_rate": 0.00039693994397784046,
      "loss": 0.4299,
      "step": 243
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7954990215264188,
      "eval_f1": 0.8168273444347064,
      "eval_loss": 0.4317159950733185,
      "eval_precision": 0.7396825396825397,
      "eval_recall": 0.9119373776908023,
      "eval_runtime": 28.0555,
      "eval_samples_per_second": 36.428,
      "eval_steps_per_second": 1.141,
      "step": 243
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.1239681243896484,
      "learning_rate": 0.0002977049579833804,
      "loss": 0.4141,
      "step": 324
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8052837573385518,
      "eval_f1": 0.7995971802618329,
      "eval_loss": 0.41590413451194763,
      "eval_precision": 0.8236514522821576,
      "eval_recall": 0.776908023483366,
      "eval_runtime": 28.4853,
      "eval_samples_per_second": 35.878,
      "eval_steps_per_second": 1.123,
      "step": 324
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.6590023040771484,
      "learning_rate": 0.00019846997198892023,
      "loss": 0.4038,
      "step": 405
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8160469667318982,
      "eval_f1": 0.8256029684601114,
      "eval_loss": 0.40745633840560913,
      "eval_precision": 0.7848324514991182,
      "eval_recall": 0.8708414872798435,
      "eval_runtime": 28.3469,
      "eval_samples_per_second": 36.053,
      "eval_steps_per_second": 1.129,
      "step": 405
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.6547752618789673,
      "learning_rate": 9.923498599446012e-05,
      "loss": 0.3966,
      "step": 486
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8111545988258317,
      "eval_f1": 0.8259693417493238,
      "eval_loss": 0.40955477952957153,
      "eval_precision": 0.7658862876254181,
      "eval_recall": 0.8962818003913894,
      "eval_runtime": 27.9254,
      "eval_samples_per_second": 36.598,
      "eval_steps_per_second": 1.146,
      "step": 486
    }
  ],
  "logging_steps": 500,
  "max_steps": 567,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 1414171183680.0,
  "train_batch_size": 38,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7591558093291033,
    "learning_rate": 0.0006946449019612209,
    "num_train_epochs": 7,
    "per_device_train_batch_size": 38,
    "temperature": 16
  }
}