{
  "best_metric": 0.2664705514907837,
  "best_model_checkpoint": "distilbert_lda_20_v1_book_qqp/checkpoint-2844",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 9954,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 3.046116590499878,
      "learning_rate": 4.9e-05,
      "loss": 0.3665,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8691318327974277,
      "eval_combined_score": 0.846567379991835,
      "eval_f1": 0.8240029271862422,
      "eval_loss": 0.2945827543735504,
      "eval_runtime": 17.6626,
      "eval_samples_per_second": 2289.021,
      "eval_steps_per_second": 8.945,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.208424091339111,
      "learning_rate": 4.8e-05,
      "loss": 0.2538,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8854563442987881,
      "eval_combined_score": 0.8665127743678859,
      "eval_f1": 0.8475692044369837,
      "eval_loss": 0.2664705514907837,
      "eval_runtime": 17.7681,
      "eval_samples_per_second": 2275.423,
      "eval_steps_per_second": 8.892,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.9439175128936768,
      "learning_rate": 4.7e-05,
      "loss": 0.1867,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8875092752906257,
      "eval_combined_score": 0.8703334830540692,
      "eval_f1": 0.8531576908175126,
      "eval_loss": 0.2858106791973114,
      "eval_runtime": 17.6175,
      "eval_samples_per_second": 2294.872,
      "eval_steps_per_second": 8.968,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.919783115386963,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1356,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8911204551075934,
      "eval_combined_score": 0.8694063281696711,
      "eval_f1": 0.8476922012317487,
      "eval_loss": 0.3080686330795288,
      "eval_runtime": 18.1311,
      "eval_samples_per_second": 2229.867,
      "eval_steps_per_second": 8.714,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.930952548980713,
      "learning_rate": 4.5e-05,
      "loss": 0.0999,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8922087558743508,
      "eval_combined_score": 0.8737844310403948,
      "eval_f1": 0.8553601062064389,
      "eval_loss": 0.35043051838874817,
      "eval_runtime": 18.0566,
      "eval_samples_per_second": 2239.069,
      "eval_steps_per_second": 8.75,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.648032188415527,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0769,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8895869403907989,
      "eval_combined_score": 0.8722835618073805,
      "eval_f1": 0.8549801832239621,
      "eval_loss": 0.37666118144989014,
      "eval_runtime": 18.0202,
      "eval_samples_per_second": 2243.594,
      "eval_steps_per_second": 8.768,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 5.479732990264893,
      "learning_rate": 4.3e-05,
      "loss": 0.0597,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.890972050457581,
      "eval_combined_score": 0.8732710842065101,
      "eval_f1": 0.8555701179554391,
      "eval_loss": 0.3907373547554016,
      "eval_runtime": 18.0243,
      "eval_samples_per_second": 2243.081,
      "eval_steps_per_second": 8.766,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "step": 9954,
      "total_flos": 1.686920659598684e+17,
      "train_loss": 0.16842511333996737,
      "train_runtime": 2559.1085,
      "train_samples_per_second": 7108.843,
      "train_steps_per_second": 27.783
    }
  ],
  "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": 1.686920659598684e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}