{
  "best_global_step": 6136,
  "best_metric": 0.7838113903999329,
  "best_model_checkpoint": "distilbert_rand_50_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": 1.2016059160232544,
      "learning_rate": 4.9e-05,
      "loss": 0.977,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5725929699439634,
      "eval_loss": 0.8987468481063843,
      "eval_runtime": 3.9718,
      "eval_samples_per_second": 2471.152,
      "eval_steps_per_second": 9.819,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8000738620758057,
      "learning_rate": 4.8e-05,
      "loss": 0.8673,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6247580234335202,
      "eval_loss": 0.837851345539093,
      "eval_runtime": 3.9603,
      "eval_samples_per_second": 2478.319,
      "eval_steps_per_second": 9.848,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.6773548126220703,
      "learning_rate": 4.7e-05,
      "loss": 0.7745,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6510443199184921,
      "eval_loss": 0.7959330677986145,
      "eval_runtime": 3.968,
      "eval_samples_per_second": 2473.51,
      "eval_steps_per_second": 9.829,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.9946660995483398,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7003,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6589913397860417,
      "eval_loss": 0.7838113903999329,
      "eval_runtime": 4.0196,
      "eval_samples_per_second": 2441.77,
      "eval_steps_per_second": 9.702,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.107696533203125,
      "learning_rate": 4.5e-05,
      "loss": 0.6348,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6669383596535915,
      "eval_loss": 0.801393985748291,
      "eval_runtime": 3.9602,
      "eval_samples_per_second": 2478.397,
      "eval_steps_per_second": 9.848,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.018315076828003,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5676,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6688741721854304,
      "eval_loss": 0.8456920981407166,
      "eval_runtime": 3.9571,
      "eval_samples_per_second": 2480.38,
      "eval_steps_per_second": 9.856,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.7117855548858643,
      "learning_rate": 4.3e-05,
      "loss": 0.5018,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6701986754966888,
      "eval_loss": 0.8797832131385803,
      "eval_runtime": 3.9907,
      "eval_samples_per_second": 2459.479,
      "eval_steps_per_second": 9.773,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.6385879516601562,
      "learning_rate": 4.2e-05,
      "loss": 0.4377,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6634742740703006,
      "eval_loss": 1.007616639137268,
      "eval_runtime": 4.0119,
      "eval_samples_per_second": 2446.467,
      "eval_steps_per_second": 9.721,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 4.2375969886779785,
      "learning_rate": 4.1e-05,
      "loss": 0.3811,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6592969943963322,
      "eval_loss": 1.1126846075057983,
      "eval_runtime": 3.9675,
      "eval_samples_per_second": 2473.828,
      "eval_steps_per_second": 9.83,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 2.340951304214108e+17,
      "train_loss": 0.6491266145889921,
      "train_runtime": 3245.7127,
      "train_samples_per_second": 6049.55,
      "train_steps_per_second": 23.631
    }
  ],
  "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
}