checkpoint-1141/trainer_state.json

{
  "best_metric": 0.3031200170516968,
  "best_model_checkpoint": "outputs/electra-nli-efl-tweeteval/checkpoint-1141",
  "epoch": 6.997323135755258,
  "global_step": 1141,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 9e-07,
      "loss": 0.439,
      "step": 163
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.745697074010327,
      "eval_f1": 0.732240056847258,
      "eval_loss": 0.3982622027397156,
      "eval_runtime": 8.9237,
      "eval_samples_per_second": 260.431,
      "eval_steps_per_second": 32.61,
      "step": 163
    },
    {
      "epoch": 2.0,
      "learning_rate": 8e-07,
      "loss": 0.3465,
      "step": 326
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7620481927710844,
      "eval_f1": 0.750740067157349,
      "eval_loss": 0.3448249399662018,
      "eval_runtime": 9.0059,
      "eval_samples_per_second": 258.053,
      "eval_steps_per_second": 32.312,
      "step": 326
    },
    {
      "epoch": 3.0,
      "learning_rate": 7e-07,
      "loss": 0.3089,
      "step": 489
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7693631669535284,
      "eval_f1": 0.7633651185887134,
      "eval_loss": 0.3303545117378235,
      "eval_runtime": 9.3737,
      "eval_samples_per_second": 247.927,
      "eval_steps_per_second": 31.044,
      "step": 489
    },
    {
      "epoch": 4.0,
      "learning_rate": 6e-07,
      "loss": 0.2916,
      "step": 652
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7839931153184165,
      "eval_f1": 0.7736843738845695,
      "eval_loss": 0.30892524123191833,
      "eval_runtime": 9.0737,
      "eval_samples_per_second": 256.125,
      "eval_steps_per_second": 32.071,
      "step": 652
    },
    {
      "epoch": 5.0,
      "learning_rate": 5e-07,
      "loss": 0.2826,
      "step": 815
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7839931153184165,
      "eval_f1": 0.7744514636272033,
      "eval_loss": 0.3055116832256317,
      "eval_runtime": 9.1262,
      "eval_samples_per_second": 254.651,
      "eval_steps_per_second": 31.886,
      "step": 815
    },
    {
      "epoch": 6.0,
      "learning_rate": 4e-07,
      "loss": 0.2747,
      "step": 978
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_f1": 0.7771340101366444,
      "eval_loss": 0.30407363176345825,
      "eval_runtime": 9.1656,
      "eval_samples_per_second": 253.557,
      "eval_steps_per_second": 31.749,
      "step": 978
    },
    {
      "epoch": 7.0,
      "learning_rate": 3e-07,
      "loss": 0.2678,
      "step": 1141
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7865748709122203,
      "eval_f1": 0.7768983029852594,
      "eval_loss": 0.3031200170516968,
      "eval_runtime": 9.6529,
      "eval_samples_per_second": 240.756,
      "eval_steps_per_second": 30.146,
      "step": 1141
    }
  ],
  "max_steps": 1630,
  "num_train_epochs": 10,
  "total_flos": 2.131107070199405e+16,
  "trial_name": null,
  "trial_params": null
}