{
  "best_metric": 0.4503476321697235,
  "best_model_checkpoint": "checkpoints/bert/roberta-base_gpt-4o-2024-05-13_gpt-4o-mini-2024-07-18_20240913_044355/checkpoint-871",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 4355,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.2520592212677,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.4781,
      "step": 871
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8025848142164782,
      "eval_f1": 0.883240970762469,
      "eval_loss": 0.4503476321697235,
      "eval_precision": 0.8292070326515967,
      "eval_recall": 0.9448078495502862,
      "eval_runtime": 28.9227,
      "eval_samples_per_second": 107.009,
      "eval_steps_per_second": 13.38,
      "step": 871
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.9907314777374268,
      "learning_rate": 1.2e-05,
      "loss": 0.4526,
      "step": 1742
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8048465266558966,
      "eval_f1": 0.8822152886115444,
      "eval_loss": 0.453591912984848,
      "eval_precision": 0.843400447427293,
      "eval_recall": 0.9247751430907605,
      "eval_runtime": 28.9179,
      "eval_samples_per_second": 107.027,
      "eval_steps_per_second": 13.383,
      "step": 1742
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.8395891189575195,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.424,
      "step": 2613
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8051696284329564,
      "eval_f1": 0.8837478311162522,
      "eval_loss": 0.4529026448726654,
      "eval_precision": 0.836191171105436,
      "eval_recall": 0.9370400654129191,
      "eval_runtime": 28.9142,
      "eval_samples_per_second": 107.041,
      "eval_steps_per_second": 13.384,
      "step": 2613
    },
    {
      "epoch": 4.0,
      "grad_norm": 11.389909744262695,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.3789,
      "step": 3484
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.802907915993538,
      "eval_f1": 0.8826471719892266,
      "eval_loss": 0.49696657061576843,
      "eval_precision": 0.8335755813953488,
      "eval_recall": 0.937857726901063,
      "eval_runtime": 28.9149,
      "eval_samples_per_second": 107.038,
      "eval_steps_per_second": 13.384,
      "step": 3484
    },
    {
      "epoch": 5.0,
      "grad_norm": 11.702038764953613,
      "learning_rate": 0.0,
      "loss": 0.3275,
      "step": 4355
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7945072697899839,
      "eval_f1": 0.8776923076923077,
      "eval_loss": 0.558699369430542,
      "eval_precision": 0.8286129266521424,
      "eval_recall": 0.9329517579721995,
      "eval_runtime": 28.913,
      "eval_samples_per_second": 107.045,
      "eval_steps_per_second": 13.385,
      "step": 4355
    },
    {
      "epoch": 5.0,
      "step": 4355,
      "total_flos": 3.66540011222016e+16,
      "train_loss": 0.4122221043408259,
      "train_runtime": 3728.8568,
      "train_samples_per_second": 37.36,
      "train_steps_per_second": 1.168
    }
  ],
  "logging_steps": 500,
  "max_steps": 4355,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3.66540011222016e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}