{
  "best_global_step": 6136,
  "best_metric": 0.8246227502822876,
  "best_model_checkpoint": "tiny_bert_rand_5_v1_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": 1.2438603639602661,
      "learning_rate": 4.9e-05,
      "loss": 0.9849,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5525216505348955,
      "eval_loss": 0.9320639967918396,
      "eval_runtime": 2.9458,
      "eval_samples_per_second": 3331.907,
      "eval_steps_per_second": 13.239,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3408972024917603,
      "learning_rate": 4.8e-05,
      "loss": 0.8957,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5971472236372899,
      "eval_loss": 0.8761903643608093,
      "eval_runtime": 2.8895,
      "eval_samples_per_second": 3396.748,
      "eval_steps_per_second": 13.497,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.6667389869689941,
      "learning_rate": 4.7e-05,
      "loss": 0.8353,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.615588385124809,
      "eval_loss": 0.8479040265083313,
      "eval_runtime": 2.867,
      "eval_samples_per_second": 3423.39,
      "eval_steps_per_second": 13.603,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.884446382522583,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7794,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6387162506367804,
      "eval_loss": 0.8246227502822876,
      "eval_runtime": 2.8779,
      "eval_samples_per_second": 3410.491,
      "eval_steps_per_second": 13.552,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.718855381011963,
      "learning_rate": 4.5e-05,
      "loss": 0.724,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6408558329088131,
      "eval_loss": 0.8253929018974304,
      "eval_runtime": 2.862,
      "eval_samples_per_second": 3429.398,
      "eval_steps_per_second": 13.627,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.2447354793548584,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6724,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6465613856342333,
      "eval_loss": 0.8509302139282227,
      "eval_runtime": 2.8713,
      "eval_samples_per_second": 3418.317,
      "eval_steps_per_second": 13.583,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.3346612453460693,
      "learning_rate": 4.3e-05,
      "loss": 0.6229,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6513499745287825,
      "eval_loss": 0.8634324073791504,
      "eval_runtime": 2.8558,
      "eval_samples_per_second": 3436.922,
      "eval_steps_per_second": 13.657,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.579292058944702,
      "learning_rate": 4.2e-05,
      "loss": 0.5766,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6483953132959756,
      "eval_loss": 0.9374760985374451,
      "eval_runtime": 2.9078,
      "eval_samples_per_second": 3375.35,
      "eval_steps_per_second": 13.412,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 2.9840760231018066,
      "learning_rate": 4.1e-05,
      "loss": 0.5324,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6448293428425879,
      "eval_loss": 0.9645715355873108,
      "eval_runtime": 2.8721,
      "eval_samples_per_second": 3417.328,
      "eval_steps_per_second": 13.579,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 9.268524907513344e+16,
      "train_loss": 0.735966525077129,
      "train_runtime": 1891.6558,
      "train_samples_per_second": 10379.848,
      "train_steps_per_second": 40.546
    }
  ],
  "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": 9.268524907513344e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}