trainer_state.json

{
  "best_metric": 0.8875705740958131,
  "best_model_checkpoint": "./stool-condition-classification\\checkpoint-100",
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 102,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "learning_rate": 0.0001803921568627451,
      "loss": 0.5897,
      "step": 10
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00016078431372549022,
      "loss": 0.479,
      "step": 20
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001411764705882353,
      "loss": 0.5314,
      "step": 30
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00012156862745098039,
      "loss": 0.5547,
      "step": 40
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00010196078431372549,
      "loss": 0.4872,
      "step": 50
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.23529411764706e-05,
      "loss": 0.505,
      "step": 60
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.274509803921569e-05,
      "loss": 0.5257,
      "step": 70
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.313725490196079e-05,
      "loss": 0.4404,
      "step": 80
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.3529411764705884e-05,
      "loss": 0.4018,
      "step": 90
    },
    {
      "epoch": 0.98,
      "learning_rate": 3.92156862745098e-06,
      "loss": 0.4071,
      "step": 100
    },
    {
      "epoch": 0.98,
      "eval_accuracy": 0.8179419525065963,
      "eval_auroc": 0.8875705740958131,
      "eval_f1": 0.7737704918032786,
      "eval_loss": 0.44151929020881653,
      "eval_runtime": 19.7299,
      "eval_samples_per_second": 19.209,
      "eval_sensitivity": 0.6629213483146067,
      "eval_specificty": 0.9552238805970149,
      "eval_steps_per_second": 2.433,
      "step": 100
    },
    {
      "epoch": 1.0,
      "step": 102,
      "total_flos": 1.25924483123712e+17,
      "train_loss": 0.4936492688515607,
      "train_runtime": 82.0698,
      "train_samples_per_second": 19.8,
      "train_steps_per_second": 1.243
    }
  ],
  "logging_steps": 10,
  "max_steps": 102,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 1.25924483123712e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}