{
  "best_metric": 0.3330647945404053,
  "best_model_checkpoint": "bert_uncased_L-4_H-512_A-8_sst2/checkpoint-792",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2112,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 8.263431549072266,
      "learning_rate": 4.9e-05,
      "loss": 0.3288,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8600917431192661,
      "eval_loss": 0.3597773015499115,
      "eval_runtime": 0.3533,
      "eval_samples_per_second": 2468.422,
      "eval_steps_per_second": 11.323,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 12.187209129333496,
      "learning_rate": 4.8e-05,
      "loss": 0.1917,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8830275229357798,
      "eval_loss": 0.3433310091495514,
      "eval_runtime": 0.3464,
      "eval_samples_per_second": 2517.025,
      "eval_steps_per_second": 11.546,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 9.220148086547852,
      "learning_rate": 4.7e-05,
      "loss": 0.1386,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8899082568807339,
      "eval_loss": 0.3330647945404053,
      "eval_runtime": 0.352,
      "eval_samples_per_second": 2476.954,
      "eval_steps_per_second": 11.362,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 7.551930904388428,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.108,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8761467889908257,
      "eval_loss": 0.403239369392395,
      "eval_runtime": 0.3511,
      "eval_samples_per_second": 2483.355,
      "eval_steps_per_second": 11.392,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.391944408416748,
      "learning_rate": 4.5e-05,
      "loss": 0.0871,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8669724770642202,
      "eval_loss": 0.448205828666687,
      "eval_runtime": 0.3509,
      "eval_samples_per_second": 2484.86,
      "eval_steps_per_second": 11.398,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 9.638006210327148,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0734,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8795871559633027,
      "eval_loss": 0.3766608238220215,
      "eval_runtime": 0.3422,
      "eval_samples_per_second": 2548.356,
      "eval_steps_per_second": 11.69,
      "step": 1584
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.6945831775665283,
      "learning_rate": 4.3e-05,
      "loss": 0.0592,
      "step": 1848
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8795871559633027,
      "eval_loss": 0.4376164376735687,
      "eval_runtime": 0.3423,
      "eval_samples_per_second": 2547.61,
      "eval_steps_per_second": 11.686,
      "step": 1848
    },
    {
      "epoch": 8.0,
      "grad_norm": 9.038667678833008,
      "learning_rate": 4.2e-05,
      "loss": 0.0509,
      "step": 2112
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8715596330275229,
      "eval_loss": 0.4948262870311737,
      "eval_runtime": 0.3501,
      "eval_samples_per_second": 2490.756,
      "eval_steps_per_second": 11.425,
      "step": 2112
    },
    {
      "epoch": 8.0,
      "step": 2112,
      "total_flos": 1.0654523282939904e+16,
      "train_loss": 0.12969460812481967,
      "train_runtime": 335.7916,
      "train_samples_per_second": 10028.393,
      "train_steps_per_second": 39.31
    }
  ],
  "logging_steps": 1,
  "max_steps": 13200,
  "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": 1.0654523282939904e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}