{
  "best_global_step": 6136,
  "best_metric": 0.7793874740600586,
  "best_model_checkpoint": "distilbert_rand_100_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": 1.1584560871124268,
      "learning_rate": 4.9e-05,
      "loss": 0.9746,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5843097300050942,
      "eval_loss": 0.8962944149971008,
      "eval_runtime": 4.2283,
      "eval_samples_per_second": 2321.283,
      "eval_steps_per_second": 9.224,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.9616583585739136,
      "learning_rate": 4.8e-05,
      "loss": 0.8662,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6220071319409067,
      "eval_loss": 0.8369470238685608,
      "eval_runtime": 3.9538,
      "eval_samples_per_second": 2482.425,
      "eval_steps_per_second": 9.864,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.9340876340866089,
      "learning_rate": 4.7e-05,
      "loss": 0.7741,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6513499745287825,
      "eval_loss": 0.793129026889801,
      "eval_runtime": 4.0195,
      "eval_samples_per_second": 2441.819,
      "eval_steps_per_second": 9.703,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.0214643478393555,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6996,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6645950076413653,
      "eval_loss": 0.7793874740600586,
      "eval_runtime": 3.973,
      "eval_samples_per_second": 2470.418,
      "eval_steps_per_second": 9.816,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.102876901626587,
      "learning_rate": 4.5e-05,
      "loss": 0.6337,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6726439123790117,
      "eval_loss": 0.7858434915542603,
      "eval_runtime": 3.9567,
      "eval_samples_per_second": 2480.575,
      "eval_steps_per_second": 9.857,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.7110612392425537,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5673,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6655119714722364,
      "eval_loss": 0.8410236835479736,
      "eval_runtime": 3.9872,
      "eval_samples_per_second": 2461.629,
      "eval_steps_per_second": 9.781,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.9554266929626465,
      "learning_rate": 4.3e-05,
      "loss": 0.5012,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.669587366276108,
      "eval_loss": 0.8851760625839233,
      "eval_runtime": 3.9674,
      "eval_samples_per_second": 2473.902,
      "eval_steps_per_second": 9.83,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 4.3863959312438965,
      "learning_rate": 4.2e-05,
      "loss": 0.439,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6619460010188487,
      "eval_loss": 0.9908820986747742,
      "eval_runtime": 4.0012,
      "eval_samples_per_second": 2453.036,
      "eval_steps_per_second": 9.747,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 4.297445774078369,
      "learning_rate": 4.1e-05,
      "loss": 0.3799,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.664696892511462,
      "eval_loss": 1.105431318283081,
      "eval_runtime": 4.0582,
      "eval_samples_per_second": 2418.571,
      "eval_steps_per_second": 9.61,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 2.340951304214108e+17,
      "train_loss": 0.6483924110851373,
      "train_runtime": 3246.0673,
      "train_samples_per_second": 6048.889,
      "train_steps_per_second": 23.629
    }
  ],
  "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": 2.340951304214108e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}