checkpoint-7/trainer_state.json

{
  "best_metric": 0.6193200349807739,
  "best_model_checkpoint": "banana-classificator/checkpoint-7",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 7,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 14.348304748535156,
      "learning_rate": 5e-05,
      "loss": 0.7006,
      "step": 1
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5,
      "eval_auc": 0.0,
      "eval_f1": 0.0,
      "eval_loss": 0.7023754119873047,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 0.077,
      "eval_samples_per_second": 25.969,
      "eval_steps_per_second": 12.984,
      "step": 1
    },
    {
      "epoch": 2.0,
      "grad_norm": 13.228832244873047,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 0.7335,
      "step": 2
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5,
      "eval_auc": 1.0,
      "eval_f1": 0.6666666666666666,
      "eval_loss": 0.7351648807525635,
      "eval_precision": 0.5,
      "eval_recall": 1.0,
      "eval_runtime": 0.0763,
      "eval_samples_per_second": 26.212,
      "eval_steps_per_second": 13.106,
      "step": 2
    },
    {
      "epoch": 3.0,
      "grad_norm": 13.188136100769043,
      "learning_rate": 3.571428571428572e-05,
      "loss": 0.7171,
      "step": 3
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5,
      "eval_auc": 0.0,
      "eval_f1": 0.6666666666666666,
      "eval_loss": 0.6849433779716492,
      "eval_precision": 0.5,
      "eval_recall": 1.0,
      "eval_runtime": 0.0698,
      "eval_samples_per_second": 28.66,
      "eval_steps_per_second": 14.33,
      "step": 3
    },
    {
      "epoch": 4.0,
      "grad_norm": 11.3985013961792,
      "learning_rate": 2.857142857142857e-05,
      "loss": 0.5575,
      "step": 4
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5,
      "eval_auc": 1.0,
      "eval_f1": 0.6666666666666666,
      "eval_loss": 0.6594089269638062,
      "eval_precision": 0.5,
      "eval_recall": 1.0,
      "eval_runtime": 0.071,
      "eval_samples_per_second": 28.165,
      "eval_steps_per_second": 14.082,
      "step": 4
    },
    {
      "epoch": 5.0,
      "grad_norm": 10.833952903747559,
      "learning_rate": 2.1428571428571428e-05,
      "loss": 0.5536,
      "step": 5
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 1.0,
      "eval_auc": 1.0,
      "eval_f1": 1.0,
      "eval_loss": 0.6393077373504639,
      "eval_precision": 1.0,
      "eval_recall": 1.0,
      "eval_runtime": 0.0675,
      "eval_samples_per_second": 29.649,
      "eval_steps_per_second": 14.824,
      "step": 5
    },
    {
      "epoch": 6.0,
      "grad_norm": 8.14575481414795,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.3808,
      "step": 6
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 1.0,
      "eval_auc": 1.0,
      "eval_f1": 1.0,
      "eval_loss": 0.6294888257980347,
      "eval_precision": 1.0,
      "eval_recall": 1.0,
      "eval_runtime": 0.0717,
      "eval_samples_per_second": 27.9,
      "eval_steps_per_second": 13.95,
      "step": 6
    },
    {
      "epoch": 7.0,
      "grad_norm": 7.075514793395996,
      "learning_rate": 7.142857142857143e-06,
      "loss": 0.327,
      "step": 7
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 1.0,
      "eval_auc": 1.0,
      "eval_f1": 1.0,
      "eval_loss": 0.6193200349807739,
      "eval_precision": 1.0,
      "eval_recall": 1.0,
      "eval_runtime": 0.0748,
      "eval_samples_per_second": 26.736,
      "eval_steps_per_second": 13.368,
      "step": 7
    }
  ],
  "logging_steps": 1,
  "max_steps": 8,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.01
      },
      "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": 147078871252992.0,
  "train_batch_size": 10,
  "trial_name": null,
  "trial_params": null
}