{
  "best_global_step": 7670,
  "best_metric": 0.7899179458618164,
  "best_model_checkpoint": "bert_base_rand_10_v1_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": 1.2034285068511963,
      "learning_rate": 4.9e-05,
      "loss": 0.9793,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5709628120224147,
      "eval_loss": 0.905933678150177,
      "eval_runtime": 6.1968,
      "eval_samples_per_second": 1583.89,
      "eval_steps_per_second": 6.294,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.4769881963729858,
      "learning_rate": 4.8e-05,
      "loss": 0.8721,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6046867040244523,
      "eval_loss": 0.8625889420509338,
      "eval_runtime": 6.1604,
      "eval_samples_per_second": 1593.237,
      "eval_steps_per_second": 6.331,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.1156179904937744,
      "learning_rate": 4.7e-05,
      "loss": 0.7892,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6435048395313296,
      "eval_loss": 0.8121190071105957,
      "eval_runtime": 6.138,
      "eval_samples_per_second": 1599.056,
      "eval_steps_per_second": 6.354,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.318585157394409,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7112,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6463576158940397,
      "eval_loss": 0.8098294734954834,
      "eval_runtime": 6.1413,
      "eval_samples_per_second": 1598.188,
      "eval_steps_per_second": 6.35,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.2038140296936035,
      "learning_rate": 4.5e-05,
      "loss": 0.6397,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.664696892511462,
      "eval_loss": 0.7899179458618164,
      "eval_runtime": 6.134,
      "eval_samples_per_second": 1600.105,
      "eval_steps_per_second": 6.358,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.720407009124756,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5684,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6694854814060112,
      "eval_loss": 0.8457050919532776,
      "eval_runtime": 6.1391,
      "eval_samples_per_second": 1598.77,
      "eval_steps_per_second": 6.353,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.6669671535491943,
      "learning_rate": 4.3e-05,
      "loss": 0.4954,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6683647478349465,
      "eval_loss": 0.8743336796760559,
      "eval_runtime": 6.1571,
      "eval_samples_per_second": 1594.101,
      "eval_steps_per_second": 6.334,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.97357177734375,
      "learning_rate": 4.2e-05,
      "loss": 0.4255,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6590932246561385,
      "eval_loss": 0.9792904257774353,
      "eval_runtime": 6.1536,
      "eval_samples_per_second": 1595.012,
      "eval_steps_per_second": 6.338,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 4.395272731781006,
      "learning_rate": 4.1e-05,
      "loss": 0.3614,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6596026490066225,
      "eval_loss": 1.150529146194458,
      "eval_runtime": 6.1549,
      "eval_samples_per_second": 1594.666,
      "eval_steps_per_second": 6.336,
      "step": 13806
    },
    {
      "epoch": 10.0,
      "grad_norm": 3.9099857807159424,
      "learning_rate": 4e-05,
      "loss": 0.3036,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6592969943963322,
      "eval_loss": 1.2116655111312866,
      "eval_runtime": 6.1479,
      "eval_samples_per_second": 1596.48,
      "eval_steps_per_second": 6.344,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "step": 15340,
      "total_flos": 5.166258268431053e+17,
      "train_loss": 0.6145827988446769,
      "train_runtime": 6373.0213,
      "train_samples_per_second": 3080.972,
      "train_steps_per_second": 12.035
    }
  ],
  "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": 5.166258268431053e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}