last-checkpoint/trainer_state.json

{
  "best_metric": 0.7986186145166447,
  "best_model_checkpoint": "/opt/ml/checkpoints/toxicity-target-type-identification-2023-02-26-12-48-26-617/model/checkpoint-1065",
  "epoch": 5.0,
  "global_step": 1775,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6902748414376322,
      "eval_f1": 0.7052333645117603,
      "eval_loss": 0.7145297527313232,
      "eval_precision": 0.7527983972979847,
      "eval_recall": 0.6902748414376322,
      "eval_runtime": 19.9276,
      "eval_samples_per_second": 47.472,
      "eval_steps_per_second": 5.972,
      "step": 355
    },
    {
      "epoch": 1.41,
      "learning_rate": 3.766830354166815e-05,
      "loss": 0.8011,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7928118393234672,
      "eval_f1": 0.7840254300809458,
      "eval_loss": 0.9930071234703064,
      "eval_precision": 0.783507252961592,
      "eval_recall": 0.7928118393234672,
      "eval_runtime": 20.1728,
      "eval_samples_per_second": 46.895,
      "eval_steps_per_second": 5.899,
      "step": 710
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.581272208641356e-05,
      "loss": 0.529,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8002114164904862,
      "eval_f1": 0.7986186145166447,
      "eval_loss": 1.4281493425369263,
      "eval_precision": 0.7989985820486601,
      "eval_recall": 0.8002114164904862,
      "eval_runtime": 17.0638,
      "eval_samples_per_second": 55.439,
      "eval_steps_per_second": 6.974,
      "step": 1065
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7727272727272727,
      "eval_f1": 0.7753471858137818,
      "eval_loss": 1.678259015083313,
      "eval_precision": 0.778801583678501,
      "eval_recall": 0.7727272727272727,
      "eval_runtime": 17.2821,
      "eval_samples_per_second": 54.739,
      "eval_steps_per_second": 6.886,
      "step": 1420
    },
    {
      "epoch": 4.23,
      "learning_rate": 3.3957140631158974e-05,
      "loss": 0.2706,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7727272727272727,
      "eval_f1": 0.7682734112208002,
      "eval_loss": 2.390378475189209,
      "eval_precision": 0.7659798393766588,
      "eval_recall": 0.7727272727272727,
      "eval_runtime": 19.0454,
      "eval_samples_per_second": 49.671,
      "eval_steps_per_second": 6.248,
      "step": 1775
    }
  ],
  "max_steps": 10650,
  "num_train_epochs": 30,
  "total_flos": 2462684273097684.0,
  "trial_name": null,
  "trial_params": null
}