{
  "best_metric": 0.7326732673267327,
  "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-16/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": 1.2602354288101196,
      "learning_rate": 6.163228300166189e-06,
      "loss": 0.7018,
      "step": 5
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.1513686180114746,
      "learning_rate": 1.2326456600332378e-05,
      "loss": 0.6815,
      "step": 10
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.21480131149292,
      "learning_rate": 1.8489684900498564e-05,
      "loss": 0.6818,
      "step": 15
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.8887011408805847,
      "learning_rate": 2.4652913200664756e-05,
      "loss": 0.6556,
      "step": 20
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.8671120405197144,
      "learning_rate": 3.081614150083095e-05,
      "loss": 0.6284,
      "step": 25
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.6801092028617859,
      "learning_rate": 3.697936980099713e-05,
      "loss": 0.6166,
      "step": 30
    },
    {
      "epoch": 0.73,
      "grad_norm": 2.9359517097473145,
      "learning_rate": 4.314259810116332e-05,
      "loss": 0.5407,
      "step": 35
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.6087604761123657,
      "learning_rate": 4.930582640132951e-05,
      "loss": 0.6755,
      "step": 40
    },
    {
      "epoch": 0.94,
      "grad_norm": 3.7170913219451904,
      "learning_rate": 5.54690547014957e-05,
      "loss": 0.6359,
      "step": 45
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.7326732673267327,
      "eval_loss": 0.7614002227783203,
      "eval_runtime": 1.3801,
      "eval_samples_per_second": 46.373,
      "eval_steps_per_second": 5.797,
      "step": 48
    }
  ],
  "logging_steps": 5,
  "max_steps": 480,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 692215198094688.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 5.916699168159541e-05,
    "per_device_train_batch_size": 4
  }
}