{
  "best_metric": 0.733329713344574,
  "best_model_checkpoint": "bert_uncased_L-4_H-512_A-8_wnli/checkpoint-3",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 18,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.648861050605774,
      "learning_rate": 4.9e-05,
      "loss": 0.712,
      "step": 3
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2535211267605634,
      "eval_loss": 0.733329713344574,
      "eval_runtime": 0.0389,
      "eval_samples_per_second": 1825.511,
      "eval_steps_per_second": 25.711,
      "step": 3
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.2163035869598389,
      "learning_rate": 4.8e-05,
      "loss": 0.7016,
      "step": 6
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.43661971830985913,
      "eval_loss": 0.7344297766685486,
      "eval_runtime": 0.0362,
      "eval_samples_per_second": 1959.904,
      "eval_steps_per_second": 27.604,
      "step": 6
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.7639871835708618,
      "learning_rate": 4.7e-05,
      "loss": 0.6925,
      "step": 9
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2112676056338028,
      "eval_loss": 0.7545115351676941,
      "eval_runtime": 0.0377,
      "eval_samples_per_second": 1883.59,
      "eval_steps_per_second": 26.529,
      "step": 9
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.9658556580543518,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6904,
      "step": 12
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.23943661971830985,
      "eval_loss": 0.7700263857841492,
      "eval_runtime": 0.0354,
      "eval_samples_per_second": 2008.4,
      "eval_steps_per_second": 28.287,
      "step": 12
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.9394496083259583,
      "learning_rate": 4.5e-05,
      "loss": 0.6891,
      "step": 15
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.18309859154929578,
      "eval_loss": 0.7776187658309937,
      "eval_runtime": 0.0388,
      "eval_samples_per_second": 1829.706,
      "eval_steps_per_second": 25.771,
      "step": 15
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.0742976665496826,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6948,
      "step": 18
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.16901408450704225,
      "eval_loss": 0.7862564921379089,
      "eval_runtime": 0.0355,
      "eval_samples_per_second": 2000.279,
      "eval_steps_per_second": 28.173,
      "step": 18
    },
    {
      "epoch": 6.0,
      "step": 18,
      "total_flos": 75342124062720.0,
      "train_loss": 0.6967346668243408,
      "train_runtime": 6.9609,
      "train_samples_per_second": 4561.214,
      "train_steps_per_second": 21.549
    }
  ],
  "logging_steps": 1,
  "max_steps": 150,
  "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": 75342124062720.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}