checkpoint-200/trainer_state.json

{
  "best_global_step": 200,
  "best_metric": 0.9974785678265254,
  "best_model_checkpoint": "nci-binary-detector/checkpoint-200",
  "epoch": 0.32679738562091504,
  "eval_steps": 100,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.032679738562091505,
      "grad_norm": 2.849823236465454,
      "learning_rate": 1.2418300653594772e-06,
      "loss": 0.1789,
      "step": 20
    },
    {
      "epoch": 0.06535947712418301,
      "grad_norm": 1.1499906778335571,
      "learning_rate": 2.549019607843137e-06,
      "loss": 0.1045,
      "step": 40
    },
    {
      "epoch": 0.09803921568627451,
      "grad_norm": 1.5816751718521118,
      "learning_rate": 3.856209150326798e-06,
      "loss": 0.0389,
      "step": 60
    },
    {
      "epoch": 0.13071895424836602,
      "grad_norm": 0.3664398789405823,
      "learning_rate": 5.163398692810458e-06,
      "loss": 0.0164,
      "step": 80
    },
    {
      "epoch": 0.16339869281045752,
      "grad_norm": 0.15663352608680725,
      "learning_rate": 6.470588235294119e-06,
      "loss": 0.0123,
      "step": 100
    },
    {
      "epoch": 0.16339869281045752,
      "eval_accuracy": 0.9837869137232195,
      "eval_f1": 0.9860279441117764,
      "eval_loss": 0.005597150418907404,
      "eval_precision": 0.9743589743589743,
      "eval_recall": 0.997979797979798,
      "eval_roc_auc": 0.9985759905705631,
      "eval_runtime": 11.5844,
      "eval_samples_per_second": 149.08,
      "eval_steps_per_second": 4.661,
      "step": 100
    },
    {
      "epoch": 0.19607843137254902,
      "grad_norm": 0.10866066068410873,
      "learning_rate": 7.77777777777778e-06,
      "loss": 0.0073,
      "step": 120
    },
    {
      "epoch": 0.22875816993464052,
      "grad_norm": 0.06542709469795227,
      "learning_rate": 9.084967320261438e-06,
      "loss": 0.0019,
      "step": 140
    },
    {
      "epoch": 0.26143790849673204,
      "grad_norm": 0.15572959184646606,
      "learning_rate": 1.03921568627451e-05,
      "loss": 0.0014,
      "step": 160
    },
    {
      "epoch": 0.29411764705882354,
      "grad_norm": 0.009457970038056374,
      "learning_rate": 1.169934640522876e-05,
      "loss": 0.0066,
      "step": 180
    },
    {
      "epoch": 0.32679738562091504,
      "grad_norm": 0.016104549169540405,
      "learning_rate": 1.3006535947712419e-05,
      "loss": 0.0007,
      "step": 200
    },
    {
      "epoch": 0.32679738562091504,
      "eval_accuracy": 0.9971048060220035,
      "eval_f1": 0.9974785678265254,
      "eval_loss": 0.0025405457708984613,
      "eval_precision": 0.9959718026183283,
      "eval_recall": 0.998989898989899,
      "eval_roc_auc": 0.998836396529748,
      "eval_runtime": 8.7693,
      "eval_samples_per_second": 196.936,
      "eval_steps_per_second": 6.158,
      "step": 200
    }
  ],
  "logging_steps": 20,
  "max_steps": 3060,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 200,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2180850214502400.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}