{
  "best_metric": 0.6611328721046448,
  "best_model_checkpoint": "hBERTv1_new_pretrain_48_emb_com_qnli/checkpoint-2457",
  "epoch": 8.0,
  "global_step": 6552,
  "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": 0.6862,
      "step": 819
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5513454146073586,
      "eval_loss": 0.6995250582695007,
      "eval_runtime": 24.6277,
      "eval_samples_per_second": 221.823,
      "eval_steps_per_second": 1.746,
      "step": 819
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 0.6677,
      "step": 1638
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.594911220940875,
      "eval_loss": 0.6624441146850586,
      "eval_runtime": 24.5823,
      "eval_samples_per_second": 222.233,
      "eval_steps_per_second": 1.749,
      "step": 1638
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 0.6553,
      "step": 2457
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5983891634632985,
      "eval_loss": 0.6611328721046448,
      "eval_runtime": 24.601,
      "eval_samples_per_second": 222.064,
      "eval_steps_per_second": 1.748,
      "step": 2457
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.6426,
      "step": 3276
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5762401610836537,
      "eval_loss": 0.6866379976272583,
      "eval_runtime": 24.809,
      "eval_samples_per_second": 220.203,
      "eval_steps_per_second": 1.733,
      "step": 3276
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 0.6244,
      "step": 4095
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5950942705473183,
      "eval_loss": 0.675081729888916,
      "eval_runtime": 24.7609,
      "eval_samples_per_second": 220.63,
      "eval_steps_per_second": 1.737,
      "step": 4095
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 0.6136,
      "step": 4914
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5727622185612301,
      "eval_loss": 0.6902745962142944,
      "eval_runtime": 24.6831,
      "eval_samples_per_second": 221.326,
      "eval_steps_per_second": 1.742,
      "step": 4914
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 0.6036,
      "step": 5733
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5731283177741168,
      "eval_loss": 0.7188088297843933,
      "eval_runtime": 24.6865,
      "eval_samples_per_second": 221.295,
      "eval_steps_per_second": 1.742,
      "step": 5733
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 0.5855,
      "step": 6552
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5824638477027274,
      "eval_loss": 0.7175068259239197,
      "eval_runtime": 24.6261,
      "eval_samples_per_second": 221.838,
      "eval_steps_per_second": 1.746,
      "step": 6552
    },
    {
      "epoch": 8.0,
      "step": 6552,
      "total_flos": 1.2422266908730982e+17,
      "train_loss": 0.6348617563026437,
      "train_runtime": 11786.3018,
      "train_samples_per_second": 444.342,
      "train_steps_per_second": 3.474
    }
  ],
  "max_steps": 40950,
  "num_train_epochs": 50,
  "total_flos": 1.2422266908730982e+17,
  "trial_name": null,
  "trial_params": null
}