{
  "best_global_step": 7670,
  "best_metric": 0.8408448696136475,
  "best_model_checkpoint": "distilbert_rand_5_v2_mnli/checkpoint-7670",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 15340,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.0643099546432495,
      "learning_rate": 4.9e-05,
      "loss": 0.9799,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5641365257259296,
      "eval_loss": 0.9140892624855042,
      "eval_runtime": 4.0713,
      "eval_samples_per_second": 2410.79,
      "eval_steps_per_second": 9.579,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5242249965667725,
      "learning_rate": 4.8e-05,
      "loss": 0.8826,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5986754966887418,
      "eval_loss": 0.8755787014961243,
      "eval_runtime": 3.947,
      "eval_samples_per_second": 2486.698,
      "eval_steps_per_second": 9.881,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.6232391595840454,
      "learning_rate": 4.7e-05,
      "loss": 0.8138,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6195618950585838,
      "eval_loss": 0.847908079624176,
      "eval_runtime": 3.9736,
      "eval_samples_per_second": 2470.039,
      "eval_steps_per_second": 9.815,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.1782569885253906,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7453,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6264900662251656,
      "eval_loss": 0.8541532158851624,
      "eval_runtime": 3.9774,
      "eval_samples_per_second": 2467.665,
      "eval_steps_per_second": 9.805,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.293247938156128,
      "learning_rate": 4.5e-05,
      "loss": 0.6741,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6393275598573612,
      "eval_loss": 0.8408448696136475,
      "eval_runtime": 3.9869,
      "eval_samples_per_second": 2461.816,
      "eval_steps_per_second": 9.782,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.5895371437072754,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6026,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6431991849210392,
      "eval_loss": 0.8902766704559326,
      "eval_runtime": 3.984,
      "eval_samples_per_second": 2463.585,
      "eval_steps_per_second": 9.789,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.590268611907959,
      "learning_rate": 4.3e-05,
      "loss": 0.533,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6423841059602649,
      "eval_loss": 0.9374910593032837,
      "eval_runtime": 3.9726,
      "eval_samples_per_second": 2470.648,
      "eval_steps_per_second": 9.817,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.947406053543091,
      "learning_rate": 4.2e-05,
      "loss": 0.4663,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6319918492103923,
      "eval_loss": 1.04335355758667,
      "eval_runtime": 3.9943,
      "eval_samples_per_second": 2457.227,
      "eval_steps_per_second": 9.764,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 3.801062822341919,
      "learning_rate": 4.1e-05,
      "loss": 0.4071,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6342333163525217,
      "eval_loss": 1.1425281763076782,
      "eval_runtime": 3.9947,
      "eval_samples_per_second": 2457.015,
      "eval_steps_per_second": 9.763,
      "step": 13806
    },
    {
      "epoch": 10.0,
      "grad_norm": 3.4782166481018066,
      "learning_rate": 4e-05,
      "loss": 0.3534,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6374936321956189,
      "eval_loss": 1.1938135623931885,
      "eval_runtime": 3.9681,
      "eval_samples_per_second": 2473.499,
      "eval_steps_per_second": 9.828,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "step": 15340,
      "total_flos": 2.6010570046823424e+17,
      "train_loss": 0.6458142256954348,
      "train_runtime": 3608.2461,
      "train_samples_per_second": 5441.73,
      "train_steps_per_second": 21.257
    }
  ],
  "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": 2.6010570046823424e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}