{
  "best_metric": 0.6921113133430481,
  "best_model_checkpoint": "bert_tiny_lda_20_v1_rte/checkpoint-20",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 70,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9686848521232605,
      "learning_rate": 4.9e-05,
      "loss": 0.7103,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4693140794223827,
      "eval_loss": 0.7053672075271606,
      "eval_runtime": 0.0931,
      "eval_samples_per_second": 2975.903,
      "eval_steps_per_second": 21.487,
      "step": 10
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5997108221054077,
      "learning_rate": 4.8e-05,
      "loss": 0.6946,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5379061371841155,
      "eval_loss": 0.6921113133430481,
      "eval_runtime": 0.1055,
      "eval_samples_per_second": 2624.727,
      "eval_steps_per_second": 18.951,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.5119341015815735,
      "learning_rate": 4.7e-05,
      "loss": 0.6878,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5090252707581228,
      "eval_loss": 0.6934157609939575,
      "eval_runtime": 0.1004,
      "eval_samples_per_second": 2759.214,
      "eval_steps_per_second": 19.922,
      "step": 30
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.572175920009613,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6758,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5667870036101083,
      "eval_loss": 0.693408727645874,
      "eval_runtime": 0.0976,
      "eval_samples_per_second": 2837.18,
      "eval_steps_per_second": 20.485,
      "step": 40
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.0510611534118652,
      "learning_rate": 4.5e-05,
      "loss": 0.6609,
      "step": 50
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5342960288808665,
      "eval_loss": 0.7402167320251465,
      "eval_runtime": 0.0988,
      "eval_samples_per_second": 2802.908,
      "eval_steps_per_second": 20.238,
      "step": 50
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.9776812195777893,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6213,
      "step": 60
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.51985559566787,
      "eval_loss": 0.7342234253883362,
      "eval_runtime": 0.0975,
      "eval_samples_per_second": 2840.648,
      "eval_steps_per_second": 20.51,
      "step": 60
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.8877984285354614,
      "learning_rate": 4.3e-05,
      "loss": 0.5385,
      "step": 70
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5379061371841155,
      "eval_loss": 0.8210722208023071,
      "eval_runtime": 0.0964,
      "eval_samples_per_second": 2872.023,
      "eval_steps_per_second": 20.737,
      "step": 70
    },
    {
      "epoch": 7.0,
      "step": 70,
      "total_flos": 457076966492160.0,
      "train_loss": 0.655616010938372,
      "train_runtime": 16.8113,
      "train_samples_per_second": 7405.714,
      "train_steps_per_second": 29.742
    }
  ],
  "logging_steps": 1,
  "max_steps": 500,
  "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": 457076966492160.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}