{
  "best_global_step": 6136,
  "best_metric": 0.8023750185966492,
  "best_model_checkpoint": "bert_base_km_10_v2_mnli/checkpoint-6136",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 13806,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.2520229816436768,
      "learning_rate": 4.9e-05,
      "loss": 0.9874,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5664798777381559,
      "eval_loss": 0.9179165959358215,
      "eval_runtime": 6.2446,
      "eval_samples_per_second": 1571.765,
      "eval_steps_per_second": 6.245,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3258386850357056,
      "learning_rate": 4.8e-05,
      "loss": 0.8724,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6133469179826796,
      "eval_loss": 0.8524137735366821,
      "eval_runtime": 6.1384,
      "eval_samples_per_second": 1598.955,
      "eval_steps_per_second": 6.353,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.7907534837722778,
      "learning_rate": 4.7e-05,
      "loss": 0.7899,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6418746816097809,
      "eval_loss": 0.8109418749809265,
      "eval_runtime": 6.1365,
      "eval_samples_per_second": 1599.447,
      "eval_steps_per_second": 6.355,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.9120267629623413,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.711,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6557310239429445,
      "eval_loss": 0.8023750185966492,
      "eval_runtime": 6.1014,
      "eval_samples_per_second": 1608.648,
      "eval_steps_per_second": 6.392,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.17927622795105,
      "learning_rate": 4.5e-05,
      "loss": 0.6313,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6573611818644931,
      "eval_loss": 0.8145239949226379,
      "eval_runtime": 6.1389,
      "eval_samples_per_second": 1598.816,
      "eval_steps_per_second": 6.353,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.629082679748535,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5479,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6604177279673968,
      "eval_loss": 0.8872759342193604,
      "eval_runtime": 6.1487,
      "eval_samples_per_second": 1596.277,
      "eval_steps_per_second": 6.343,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.794588327407837,
      "learning_rate": 4.3e-05,
      "loss": 0.4646,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.660825267447784,
      "eval_loss": 0.9554475545883179,
      "eval_runtime": 6.1545,
      "eval_samples_per_second": 1594.77,
      "eval_steps_per_second": 6.337,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.316474199295044,
      "learning_rate": 4.2e-05,
      "loss": 0.386,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6523688232297504,
      "eval_loss": 1.0366873741149902,
      "eval_runtime": 6.1465,
      "eval_samples_per_second": 1596.846,
      "eval_steps_per_second": 6.345,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 4.24645471572876,
      "learning_rate": 4.1e-05,
      "loss": 0.3153,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6495160468670402,
      "eval_loss": 1.245375394821167,
      "eval_runtime": 6.158,
      "eval_samples_per_second": 1593.851,
      "eval_steps_per_second": 6.333,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 4.6496324415879475e+17,
      "train_loss": 0.6339628707839322,
      "train_runtime": 5714.6331,
      "train_samples_per_second": 3435.934,
      "train_steps_per_second": 13.422
    }
  ],
  "logging_steps": 1,
  "max_steps": 76700,
  "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": 4.6496324415879475e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}