{
  "best_global_step": 93,
  "best_metric": 0.9518677022450607,
  "best_model_checkpoint": "lora_pii_detector_mmbert-32k_r8_token_model/checkpoint-93",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 93,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3225806451612903,
      "grad_norm": 28.529237747192383,
      "learning_rate": 2.7000000000000002e-05,
      "loss": 5.3341,
      "step": 10
    },
    {
      "epoch": 0.6451612903225806,
      "grad_norm": 28.822555541992188,
      "learning_rate": 2.9715728395293587e-05,
      "loss": 4.263,
      "step": 20
    },
    {
      "epoch": 0.967741935483871,
      "grad_norm": 23.09440803527832,
      "learning_rate": 2.874688621590339e-05,
      "loss": 3.1418,
      "step": 30
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5290275761973875,
      "eval_f1": 0.6812475482822514,
      "eval_loss": 1.713678240776062,
      "eval_precision": 0.9564547130584865,
      "eval_recall": 0.5290275761973876,
      "eval_runtime": 0.6046,
      "eval_samples_per_second": 203.455,
      "eval_steps_per_second": 13.233,
      "step": 31
    },
    {
      "epoch": 1.2903225806451613,
      "grad_norm": 14.115083694458008,
      "learning_rate": 2.7135254915624213e-05,
      "loss": 2.1739,
      "step": 40
    },
    {
      "epoch": 1.6129032258064515,
      "grad_norm": 9.12806224822998,
      "learning_rate": 2.495619258285757e-05,
      "loss": 1.5243,
      "step": 50
    },
    {
      "epoch": 1.935483870967742,
      "grad_norm": 8.78358268737793,
      "learning_rate": 2.231158975042979e-05,
      "loss": 1.3747,
      "step": 60
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.944121915820029,
      "eval_f1": 0.9465152764691677,
      "eval_loss": 0.27175939083099365,
      "eval_precision": 0.9492522125342383,
      "eval_recall": 0.944121915820029,
      "eval_runtime": 0.3095,
      "eval_samples_per_second": 397.43,
      "eval_steps_per_second": 25.849,
      "step": 62
    },
    {
      "epoch": 2.258064516129032,
      "grad_norm": 4.098817348480225,
      "learning_rate": 1.9325105107520264e-05,
      "loss": 1.1873,
      "step": 70
    },
    {
      "epoch": 2.5806451612903225,
      "grad_norm": 7.473396301269531,
      "learning_rate": 1.613638335679216e-05,
      "loss": 1.0562,
      "step": 80
    },
    {
      "epoch": 2.903225806451613,
      "grad_norm": 8.857183456420898,
      "learning_rate": 1.2894525583299835e-05,
      "loss": 1.1235,
      "step": 90
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9477503628447025,
      "eval_f1": 0.9518677022450607,
      "eval_loss": 0.24011534452438354,
      "eval_precision": 0.9565055622257022,
      "eval_recall": 0.9477503628447025,
      "eval_runtime": 0.2977,
      "eval_samples_per_second": 413.23,
      "eval_steps_per_second": 26.877,
      "step": 93
    }
  ],
  "logging_steps": 10,
  "max_steps": 155,
  "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": 101358543358800.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}