{
  "best_global_step": 746,
  "best_metric": 2.618539571762085,
  "best_model_checkpoint": "./gpt2_conv/checkpoint-746",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2611,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.6198129653930664,
      "learning_rate": 0.0009501340482573728,
      "loss": 2.6864,
      "step": 373
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.6242876052856445,
      "eval_runtime": 7.6714,
      "eval_samples_per_second": 97.114,
      "eval_steps_per_second": 12.253,
      "step": 373
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.9966814517974854,
      "learning_rate": 0.0009001340482573726,
      "loss": 2.5097,
      "step": 746
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.618539571762085,
      "eval_runtime": 5.9683,
      "eval_samples_per_second": 124.826,
      "eval_steps_per_second": 15.75,
      "step": 746
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.020508050918579,
      "learning_rate": 0.0008501340482573726,
      "loss": 2.3432,
      "step": 1119
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.628633499145508,
      "eval_runtime": 5.8711,
      "eval_samples_per_second": 126.893,
      "eval_steps_per_second": 16.011,
      "step": 1119
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.9825632572174072,
      "learning_rate": 0.0008001340482573727,
      "loss": 2.1637,
      "step": 1492
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.6619577407836914,
      "eval_runtime": 5.8085,
      "eval_samples_per_second": 128.261,
      "eval_steps_per_second": 16.183,
      "step": 1492
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.038879156112671,
      "learning_rate": 0.0007501340482573727,
      "loss": 2.0068,
      "step": 1865
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.7242045402526855,
      "eval_runtime": 5.8714,
      "eval_samples_per_second": 126.886,
      "eval_steps_per_second": 16.01,
      "step": 1865
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.051107883453369,
      "learning_rate": 0.0007001340482573728,
      "loss": 1.8635,
      "step": 2238
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.774700880050659,
      "eval_runtime": 5.9959,
      "eval_samples_per_second": 124.252,
      "eval_steps_per_second": 15.677,
      "step": 2238
    },
    {
      "epoch": 7.0,
      "grad_norm": 6.0618062019348145,
      "learning_rate": 0.0006501340482573726,
      "loss": 1.7221,
      "step": 2611
    },
    {
      "epoch": 7.0,
      "eval_loss": 2.907008409500122,
      "eval_runtime": 5.7642,
      "eval_samples_per_second": 129.246,
      "eval_steps_per_second": 16.308,
      "step": 2611
    }
  ],
  "logging_steps": 1,
  "max_steps": 7460,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 1120715234082816.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}