{
  "best_metric": 0.6285243252934658,
  "best_model_checkpoint": "students_scores_model/run-1/checkpoint-1128",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 1128,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_f1": 0.5135559583730447,
      "eval_loss": 1.0615456104278564,
      "eval_runtime": 37.678,
      "eval_samples_per_second": 59.716,
      "eval_steps_per_second": 7.484,
      "step": 282
    },
    {
      "epoch": 1.773049645390071,
      "grad_norm": 11.819360733032227,
      "learning_rate": 8.537794225405964e-06,
      "loss": 1.0289,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.5853742204012189,
      "eval_loss": 0.9459031224250793,
      "eval_runtime": 37.7163,
      "eval_samples_per_second": 59.656,
      "eval_steps_per_second": 7.477,
      "step": 564
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.6179112498233506,
      "eval_loss": 0.8894566893577576,
      "eval_runtime": 37.6552,
      "eval_samples_per_second": 59.753,
      "eval_steps_per_second": 7.489,
      "step": 846
    },
    {
      "epoch": 3.546099290780142,
      "grad_norm": 19.25679588317871,
      "learning_rate": 3.8466984971609285e-06,
      "loss": 0.8051,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.6285243252934658,
      "eval_loss": 0.8603602647781372,
      "eval_runtime": 37.769,
      "eval_samples_per_second": 59.573,
      "eval_steps_per_second": 7.466,
      "step": 1128
    }
  ],
  "logging_steps": 500,
  "max_steps": 1410,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8400122116761600.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.3228889953650998e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 32,
    "weight_decay": 0.0004860339799795757
  }
}