{
  "best_metric": 0.9361986026141007,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-swiss_re_assignment-bodypart/run-7/checkpoint-3128",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 3128,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.64,
      "grad_norm": 3.043917655944824,
      "learning_rate": 1.1523885830392892e-05,
      "loss": 1.301,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8952649216735153,
      "eval_loss": 0.44891732931137085,
      "eval_runtime": 4.2374,
      "eval_samples_per_second": 184.781,
      "eval_steps_per_second": 3.068,
      "step": 782
    },
    {
      "epoch": 1.28,
      "grad_norm": 46.61455154418945,
      "learning_rate": 9.83416650042326e-06,
      "loss": 0.4247,
      "step": 1000
    },
    {
      "epoch": 1.92,
      "grad_norm": 36.788780212402344,
      "learning_rate": 8.144447170453627e-06,
      "loss": 0.2417,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9069869583316635,
      "eval_loss": 0.41448912024497986,
      "eval_runtime": 5.7752,
      "eval_samples_per_second": 135.58,
      "eval_steps_per_second": 2.251,
      "step": 1564
    },
    {
      "epoch": 2.56,
      "grad_norm": 0.9252055287361145,
      "learning_rate": 6.454727840483996e-06,
      "loss": 0.2104,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9338136133834791,
      "eval_loss": 0.3658573627471924,
      "eval_runtime": 4.3869,
      "eval_samples_per_second": 178.488,
      "eval_steps_per_second": 2.963,
      "step": 2346
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.9745433926582336,
      "learning_rate": 4.765008510514363e-06,
      "loss": 0.171,
      "step": 2500
    },
    {
      "epoch": 3.84,
      "grad_norm": 0.03158050402998924,
      "learning_rate": 3.075289180544731e-06,
      "loss": 0.1593,
      "step": 3000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9361986026141007,
      "eval_loss": 0.3598783314228058,
      "eval_runtime": 4.2759,
      "eval_samples_per_second": 183.12,
      "eval_steps_per_second": 3.04,
      "step": 3128
    }
  ],
  "logging_steps": 500,
  "max_steps": 3910,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 140988338859072.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.3213605160362524e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 4,
    "seed": 10
  }
}