trainer_state.json

{
  "best_metric": 0.2688056230545044,
  "best_model_checkpoint": "distilbert_lda_100_v1_book_qqp/checkpoint-4266",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 11376,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.6672399044036865,
      "learning_rate": 4.9e-05,
      "loss": 0.3709,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8678456591639871,
      "eval_combined_score": 0.8435809405691093,
      "eval_f1": 0.8193162219742315,
      "eval_loss": 0.3001379072666168,
      "eval_runtime": 12.6438,
      "eval_samples_per_second": 3197.621,
      "eval_steps_per_second": 12.496,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.6408467292785645,
      "learning_rate": 4.8e-05,
      "loss": 0.2573,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8761810536730151,
      "eval_combined_score": 0.8594045790213551,
      "eval_f1": 0.842628104369695,
      "eval_loss": 0.28343185782432556,
      "eval_runtime": 12.6843,
      "eval_samples_per_second": 3187.412,
      "eval_steps_per_second": 12.456,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.3231992721557617,
      "learning_rate": 4.7e-05,
      "loss": 0.1922,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8875092752906257,
      "eval_combined_score": 0.8687645307992504,
      "eval_f1": 0.850019786307875,
      "eval_loss": 0.2688056230545044,
      "eval_runtime": 12.5615,
      "eval_samples_per_second": 3218.574,
      "eval_steps_per_second": 12.578,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.48140811920166,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1411,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8910462527825872,
      "eval_combined_score": 0.8708443890637921,
      "eval_f1": 0.8506425253449971,
      "eval_loss": 0.3128622770309448,
      "eval_runtime": 13.017,
      "eval_samples_per_second": 3105.941,
      "eval_steps_per_second": 12.138,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.0898919105529785,
      "learning_rate": 4.5e-05,
      "loss": 0.105,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8932475884244373,
      "eval_combined_score": 0.8747814611674087,
      "eval_f1": 0.8563153339103802,
      "eval_loss": 0.325675368309021,
      "eval_runtime": 12.5089,
      "eval_samples_per_second": 3232.109,
      "eval_steps_per_second": 12.631,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 5.714954853057861,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0794,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8901063566658421,
      "eval_combined_score": 0.8723434607929086,
      "eval_f1": 0.8545805649199751,
      "eval_loss": 0.36958828568458557,
      "eval_runtime": 12.5087,
      "eval_samples_per_second": 3232.151,
      "eval_steps_per_second": 12.631,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 6.753868579864502,
      "learning_rate": 4.3e-05,
      "loss": 0.0646,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8888943853574078,
      "eval_combined_score": 0.8712875835581828,
      "eval_f1": 0.8536807817589577,
      "eval_loss": 0.3887239396572113,
      "eval_runtime": 12.5806,
      "eval_samples_per_second": 3213.679,
      "eval_steps_per_second": 12.559,
      "step": 9954
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.0667166709899902,
      "learning_rate": 4.2e-05,
      "loss": 0.0528,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.89060103883255,
      "eval_combined_score": 0.873538180447524,
      "eval_f1": 0.856475322062498,
      "eval_loss": 0.42925697565078735,
      "eval_runtime": 12.6223,
      "eval_samples_per_second": 3203.05,
      "eval_steps_per_second": 12.517,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "step": 11376,
      "total_flos": 1.927909325255639e+17,
      "train_loss": 0.1579339621606926,
      "train_runtime": 2009.8047,
      "train_samples_per_second": 9051.775,
      "train_steps_per_second": 35.377
    }
  ],
  "logging_steps": 1,
  "max_steps": 71100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.927909325255639e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}