{
  "best_metric": 0.6672190427780151,
  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init__qnli/checkpoint-4914",
  "epoch": 11.0,
  "global_step": 9009,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 0.6909,
      "step": 819
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5652571846970529,
      "eval_loss": 0.6782859563827515,
      "eval_runtime": 9.8372,
      "eval_samples_per_second": 555.34,
      "eval_steps_per_second": 4.371,
      "step": 819
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 0.684,
      "step": 1638
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5099762035511624,
      "eval_loss": 0.6904051899909973,
      "eval_runtime": 9.816,
      "eval_samples_per_second": 556.542,
      "eval_steps_per_second": 4.381,
      "step": 1638
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 0.6765,
      "step": 2457
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5881383855024712,
      "eval_loss": 0.6709150671958923,
      "eval_runtime": 9.8166,
      "eval_samples_per_second": 556.504,
      "eval_steps_per_second": 4.38,
      "step": 2457
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.6696,
      "step": 3276
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5694673256452498,
      "eval_loss": 0.6773616671562195,
      "eval_runtime": 9.8364,
      "eval_samples_per_second": 555.383,
      "eval_steps_per_second": 4.371,
      "step": 3276
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 0.6676,
      "step": 4095
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5903349807797913,
      "eval_loss": 0.6704134345054626,
      "eval_runtime": 9.8333,
      "eval_samples_per_second": 555.562,
      "eval_steps_per_second": 4.373,
      "step": 4095
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 0.6626,
      "step": 4914
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.598572213069742,
      "eval_loss": 0.6672190427780151,
      "eval_runtime": 9.8615,
      "eval_samples_per_second": 553.973,
      "eval_steps_per_second": 4.36,
      "step": 4914
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 0.6661,
      "step": 5733
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5907010799926781,
      "eval_loss": 0.6703054904937744,
      "eval_runtime": 9.8372,
      "eval_samples_per_second": 555.339,
      "eval_steps_per_second": 4.371,
      "step": 5733
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 0.6642,
      "step": 6552
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5960095185795351,
      "eval_loss": 0.6692940592765808,
      "eval_runtime": 9.8258,
      "eval_samples_per_second": 555.985,
      "eval_steps_per_second": 4.376,
      "step": 6552
    },
    {
      "epoch": 9.0,
      "learning_rate": 3.28e-05,
      "loss": 0.6698,
      "step": 7371
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5799011532125206,
      "eval_loss": 0.6733358502388,
      "eval_runtime": 9.846,
      "eval_samples_per_second": 554.842,
      "eval_steps_per_second": 4.367,
      "step": 7371
    },
    {
      "epoch": 10.0,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.6724,
      "step": 8190
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5636097382390628,
      "eval_loss": 0.6815194487571716,
      "eval_runtime": 9.8244,
      "eval_samples_per_second": 556.067,
      "eval_steps_per_second": 4.377,
      "step": 8190
    },
    {
      "epoch": 11.0,
      "learning_rate": 3.1200000000000006e-05,
      "loss": 0.68,
      "step": 9009
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.5427420831045213,
      "eval_loss": 0.6908370852470398,
      "eval_runtime": 9.8206,
      "eval_samples_per_second": 556.277,
      "eval_steps_per_second": 4.379,
      "step": 9009
    },
    {
      "epoch": 11.0,
      "step": 9009,
      "total_flos": 1.706331010666332e+17,
      "train_loss": 0.6730744434601856,
      "train_runtime": 6372.3482,
      "train_samples_per_second": 821.856,
      "train_steps_per_second": 6.426
    }
  ],
  "max_steps": 40950,
  "num_train_epochs": 50,
  "total_flos": 1.706331010666332e+17,
  "trial_name": null,
  "trial_params": null
}