{
  "best_global_step": 10,
  "best_metric": 0.6952561140060425,
  "best_model_checkpoint": "bert_base_km_100_v2_rte/checkpoint-10",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.3114423751831055,
      "learning_rate": 4.9e-05,
      "loss": 0.7065,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5126353790613718,
      "eval_loss": 0.6952561140060425,
      "eval_runtime": 0.1897,
      "eval_samples_per_second": 1459.974,
      "eval_steps_per_second": 10.541,
      "step": 10
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.0168451070785522,
      "learning_rate": 4.8e-05,
      "loss": 0.6522,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.7045880556106567,
      "eval_runtime": 0.2391,
      "eval_samples_per_second": 1158.41,
      "eval_steps_per_second": 8.364,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.0036348104476929,
      "learning_rate": 4.7e-05,
      "loss": 0.6078,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4981949458483754,
      "eval_loss": 0.7145405411720276,
      "eval_runtime": 0.1995,
      "eval_samples_per_second": 1388.258,
      "eval_steps_per_second": 10.024,
      "step": 30
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.1682407855987549,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5363,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5342960288808665,
      "eval_loss": 0.7521734833717346,
      "eval_runtime": 0.1971,
      "eval_samples_per_second": 1405.13,
      "eval_steps_per_second": 10.145,
      "step": 40
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.5133919715881348,
      "learning_rate": 4.5e-05,
      "loss": 0.4366,
      "step": 50
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5487364620938628,
      "eval_loss": 0.8362495303153992,
      "eval_runtime": 0.1975,
      "eval_samples_per_second": 1402.249,
      "eval_steps_per_second": 10.125,
      "step": 50
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.0950589179992676,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3284,
      "step": 60
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4693140794223827,
      "eval_loss": 1.0345640182495117,
      "eval_runtime": 0.1915,
      "eval_samples_per_second": 1446.353,
      "eval_steps_per_second": 10.443,
      "step": 60
    },
    {
      "epoch": 6.0,
      "step": 60,
      "total_flos": 1965439583539200.0,
      "train_loss": 0.5446344256401062,
      "train_runtime": 38.288,
      "train_samples_per_second": 3251.673,
      "train_steps_per_second": 13.059
    }
  ],
  "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": 1965439583539200.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}