{
  "best_metric": 0.27154991030693054,
  "best_model_checkpoint": "bert_base_lda_5_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": 2.695286750793457,
      "learning_rate": 4.9e-05,
      "loss": 0.3439,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8779371753648281,
      "eval_combined_score": 0.854934258765069,
      "eval_f1": 0.8319313421653101,
      "eval_loss": 0.27763593196868896,
      "eval_runtime": 25.9261,
      "eval_samples_per_second": 1559.435,
      "eval_steps_per_second": 6.094,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.2603042125701904,
      "learning_rate": 4.8e-05,
      "loss": 0.2315,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.883007667573584,
      "eval_combined_score": 0.8676345184444632,
      "eval_f1": 0.8522613693153424,
      "eval_loss": 0.27154991030693054,
      "eval_runtime": 25.9752,
      "eval_samples_per_second": 1556.482,
      "eval_steps_per_second": 6.083,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.0674655437469482,
      "learning_rate": 4.7e-05,
      "loss": 0.1614,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8950284442245857,
      "eval_combined_score": 0.8786002184749845,
      "eval_f1": 0.8621719927253833,
      "eval_loss": 0.27375757694244385,
      "eval_runtime": 25.9504,
      "eval_samples_per_second": 1557.971,
      "eval_steps_per_second": 6.089,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.1239495277404785,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1118,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8937422705911452,
      "eval_combined_score": 0.876134126799228,
      "eval_f1": 0.8585259830073109,
      "eval_loss": 0.3073498606681824,
      "eval_runtime": 25.8301,
      "eval_samples_per_second": 1565.231,
      "eval_steps_per_second": 6.117,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.931047797203064,
      "learning_rate": 4.5e-05,
      "loss": 0.0815,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8996289883749691,
      "eval_combined_score": 0.8824595612436523,
      "eval_f1": 0.8652901341123356,
      "eval_loss": 0.34702029824256897,
      "eval_runtime": 25.9432,
      "eval_samples_per_second": 1558.403,
      "eval_steps_per_second": 6.09,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.7406182289123535,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0631,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8963146178580262,
      "eval_combined_score": 0.8799503450500674,
      "eval_f1": 0.8635860722421087,
      "eval_loss": 0.3771444261074066,
      "eval_runtime": 25.9058,
      "eval_samples_per_second": 1560.656,
      "eval_steps_per_second": 6.099,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.645276427268982,
      "learning_rate": 4.3e-05,
      "loss": 0.0515,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8971555775414296,
      "eval_combined_score": 0.8802300999022816,
      "eval_f1": 0.8633046222631338,
      "eval_loss": 0.3933800160884857,
      "eval_runtime": 25.775,
      "eval_samples_per_second": 1568.573,
      "eval_steps_per_second": 6.13,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "step": 9954,
      "total_flos": 3.3506166766980096e+17,
      "train_loss": 0.14925767140908727,
      "train_runtime": 4316.4517,
      "train_samples_per_second": 4214.642,
      "train_steps_per_second": 16.472
    }
  ],
  "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": 3.3506166766980096e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}