{
  "best_global_step": 6136,
  "best_metric": 0.8286071419715881,
  "best_model_checkpoint": "distilbert_rand_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": 1.13567054271698,
      "learning_rate": 4.9e-05,
      "loss": 0.9802,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.562506367804381,
      "eval_loss": 0.9160120487213135,
      "eval_runtime": 3.9814,
      "eval_samples_per_second": 2465.202,
      "eval_steps_per_second": 9.796,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.7938745021820068,
      "learning_rate": 4.8e-05,
      "loss": 0.8841,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5992868059093225,
      "eval_loss": 0.8713922500610352,
      "eval_runtime": 4.0325,
      "eval_samples_per_second": 2433.976,
      "eval_steps_per_second": 9.671,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.9012582302093506,
      "learning_rate": 4.7e-05,
      "loss": 0.8077,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6295466123280693,
      "eval_loss": 0.8316766619682312,
      "eval_runtime": 3.991,
      "eval_samples_per_second": 2459.265,
      "eval_steps_per_second": 9.772,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.2879834175109863,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7338,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6380030565461029,
      "eval_loss": 0.8286071419715881,
      "eval_runtime": 3.9856,
      "eval_samples_per_second": 2462.607,
      "eval_steps_per_second": 9.785,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.2661197185516357,
      "learning_rate": 4.5e-05,
      "loss": 0.6635,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6465613856342333,
      "eval_loss": 0.837253749370575,
      "eval_runtime": 3.9957,
      "eval_samples_per_second": 2456.392,
      "eval_steps_per_second": 9.76,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.638463258743286,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5959,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6497198166072339,
      "eval_loss": 0.8726046085357666,
      "eval_runtime": 3.979,
      "eval_samples_per_second": 2466.669,
      "eval_steps_per_second": 9.801,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.7800238132476807,
      "learning_rate": 4.3e-05,
      "loss": 0.5281,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6498217014773306,
      "eval_loss": 0.9041654467582703,
      "eval_runtime": 4.0103,
      "eval_samples_per_second": 2447.418,
      "eval_steps_per_second": 9.725,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.5348401069641113,
      "learning_rate": 4.2e-05,
      "loss": 0.4645,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6376974019358125,
      "eval_loss": 1.0314393043518066,
      "eval_runtime": 3.9931,
      "eval_samples_per_second": 2457.995,
      "eval_steps_per_second": 9.767,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 3.6947736740112305,
      "learning_rate": 4.1e-05,
      "loss": 0.406,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6375955170657157,
      "eval_loss": 1.1399098634719849,
      "eval_runtime": 3.9747,
      "eval_samples_per_second": 2469.364,
      "eval_steps_per_second": 9.812,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 2.340951304214108e+17,
      "train_loss": 0.6737569173177084,
      "train_runtime": 3246.9896,
      "train_samples_per_second": 6047.171,
      "train_steps_per_second": 23.622
    }
  ],
  "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
}