{
  "best_global_step": 102,
  "best_metric": 0.6160990595817566,
  "best_model_checkpoint": "tiny_bert_km_5_v1_cola/checkpoint-102",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 272,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.8054304122924805,
      "learning_rate": 4.9e-05,
      "loss": 0.6163,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6178382039070129,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3876,
      "eval_samples_per_second": 2691.129,
      "eval_steps_per_second": 12.901,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.0518122911453247,
      "learning_rate": 4.8e-05,
      "loss": 0.603,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.690316379070282,
      "eval_loss": 0.6192642450332642,
      "eval_matthews_correlation": -0.020702674026557004,
      "eval_runtime": 0.3543,
      "eval_samples_per_second": 2943.468,
      "eval_steps_per_second": 14.111,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.3391984701156616,
      "learning_rate": 4.7e-05,
      "loss": 0.5893,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6893576383590698,
      "eval_loss": 0.6160990595817566,
      "eval_matthews_correlation": 0.03026012901076368,
      "eval_runtime": 0.3141,
      "eval_samples_per_second": 3320.172,
      "eval_steps_per_second": 15.916,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.786250352859497,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5577,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6778523325920105,
      "eval_loss": 0.6265153884887695,
      "eval_matthews_correlation": 0.06018138730277248,
      "eval_runtime": 0.3221,
      "eval_samples_per_second": 3237.665,
      "eval_steps_per_second": 15.521,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.1093223094940186,
      "learning_rate": 4.5e-05,
      "loss": 0.5097,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6500479578971863,
      "eval_loss": 0.6461333632469177,
      "eval_matthews_correlation": 0.09579713367112705,
      "eval_runtime": 0.3238,
      "eval_samples_per_second": 3220.948,
      "eval_steps_per_second": 15.441,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.3399136066436768,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.463,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6490891575813293,
      "eval_loss": 0.7003764510154724,
      "eval_matthews_correlation": 0.09237903909821421,
      "eval_runtime": 0.3396,
      "eval_samples_per_second": 3070.876,
      "eval_steps_per_second": 14.721,
      "step": 204
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.1917834281921387,
      "learning_rate": 4.3e-05,
      "loss": 0.4194,
      "step": 238
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6644295454025269,
      "eval_loss": 0.7483174800872803,
      "eval_matthews_correlation": 0.1263354953107224,
      "eval_runtime": 0.3085,
      "eval_samples_per_second": 3380.389,
      "eval_steps_per_second": 16.205,
      "step": 238
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.220294952392578,
      "learning_rate": 4.2e-05,
      "loss": 0.379,
      "step": 272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6251198649406433,
      "eval_loss": 0.7767397165298462,
      "eval_matthews_correlation": 0.10103917683294436,
      "eval_runtime": 0.3629,
      "eval_samples_per_second": 2873.907,
      "eval_steps_per_second": 13.777,
      "step": 272
    },
    {
      "epoch": 8.0,
      "step": 272,
      "total_flos": 1793902531485696.0,
      "train_loss": 0.5171588974840501,
      "train_runtime": 46.5693,
      "train_samples_per_second": 9180.945,
      "train_steps_per_second": 36.505
    }
  ],
  "logging_steps": 1,
  "max_steps": 1700,
  "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": 1793902531485696.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}