{
  "best_metric": 0.7326732673267327,
  "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-17/checkpoint-48",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 48,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 1.2570356130599976,
      "learning_rate": 7.514138837771986e-06,
      "loss": 0.7017,
      "step": 5
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.132084846496582,
      "learning_rate": 1.5028277675543972e-05,
      "loss": 0.678,
      "step": 10
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.1500951051712036,
      "learning_rate": 2.2542416513315956e-05,
      "loss": 0.6796,
      "step": 15
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.8457872867584229,
      "learning_rate": 3.0056555351087943e-05,
      "loss": 0.6488,
      "step": 20
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.7979240417480469,
      "learning_rate": 3.757069418885993e-05,
      "loss": 0.6218,
      "step": 25
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.605031430721283,
      "learning_rate": 4.508483302663191e-05,
      "loss": 0.612,
      "step": 30
    },
    {
      "epoch": 0.73,
      "grad_norm": 3.2487857341766357,
      "learning_rate": 5.2598971864403895e-05,
      "loss": 0.5324,
      "step": 35
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.5509129762649536,
      "learning_rate": 6.0113110702175886e-05,
      "loss": 0.6915,
      "step": 40
    },
    {
      "epoch": 0.94,
      "grad_norm": 3.975639581680298,
      "learning_rate": 6.762724953994786e-05,
      "loss": 0.643,
      "step": 45
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.7643775939941406,
      "eval_runtime": 1.3793,
      "eval_samples_per_second": 46.401,
      "eval_steps_per_second": 5.8,
      "step": 48
    }
  ],
  "logging_steps": 5,
  "max_steps": 480,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 670686130935120.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 7.213573284261106e-05,
    "per_device_train_batch_size": 4
  }
}