trainer_state.json

{
  "best_metric": 0.4279138921539274,
  "best_model_checkpoint": "trained_models/training_1691462853_boss_sentiment_t5-large/model/checkpoint-3750",
  "epoch": 2.0,
  "global_step": 3750,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.27,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 7.3573,
      "step": 500
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 0.1655,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 4e-05,
      "loss": 0.1371,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_acc": 0.904559942422373,
      "eval_f1": 0.42447092372585843,
      "eval_loss": 0.09567338228225708,
      "eval_runtime": 2233.5031,
      "eval_samples_per_second": 17.418,
      "eval_steps_per_second": 2.177,
      "step": 1875
    },
    {
      "epoch": 1.07,
      "learning_rate": 5.333333333333333e-05,
      "loss": 0.1311,
      "step": 2000
    },
    {
      "epoch": 1.33,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.1154,
      "step": 2500
    },
    {
      "epoch": 1.6,
      "learning_rate": 8e-05,
      "loss": 0.124,
      "step": 3000
    },
    {
      "epoch": 1.87,
      "learning_rate": 9.333333333333334e-05,
      "loss": 0.1268,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "eval_acc": 0.9115258071149496,
      "eval_f1": 0.4279138921539274,
      "eval_loss": 0.09571153670549393,
      "eval_runtime": 2240.719,
      "eval_samples_per_second": 17.362,
      "eval_steps_per_second": 2.17,
      "step": 3750
    }
  ],
  "max_steps": 37500,
  "num_train_epochs": 20,
  "total_flos": 1.29898507665408e+17,
  "trial_name": null,
  "trial_params": null
}