run-1/checkpoint-16/trainer_state.json

{
  "best_metric": 0.7326732673267327,
  "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-1/checkpoint-16",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 16,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.31,
      "grad_norm": 0.8568354845046997,
      "learning_rate": 2.8711716200450114e-06,
      "loss": 0.6977,
      "step": 5
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.7982807755470276,
      "learning_rate": 5.742343240090023e-06,
      "loss": 0.6961,
      "step": 10
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.32968151569366455,
      "learning_rate": 8.613514860135034e-06,
      "loss": 0.6801,
      "step": 15
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.6840133666992188,
      "eval_runtime": 1.3445,
      "eval_samples_per_second": 47.602,
      "eval_steps_per_second": 5.95,
      "step": 16
    }
  ],
  "logging_steps": 5,
  "max_steps": 160,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 812493588327408.0,
  "train_batch_size": 12,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 9.187749184144037e-06,
    "per_device_train_batch_size": 12
  }
}