{
  "best_global_step": 102,
  "best_metric": 0.6194837689399719,
  "best_model_checkpoint": "tiny_bert_km_10_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": 0.6989911794662476,
      "learning_rate": 4.9e-05,
      "loss": 0.6199,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6198849678039551,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3232,
      "eval_samples_per_second": 3227.559,
      "eval_steps_per_second": 15.472,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9705714583396912,
      "learning_rate": 4.8e-05,
      "loss": 0.6047,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.62193363904953,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3412,
      "eval_samples_per_second": 3057.25,
      "eval_steps_per_second": 14.656,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.1057320833206177,
      "learning_rate": 4.7e-05,
      "loss": 0.5929,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6194837689399719,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.3105,
      "eval_samples_per_second": 3359.633,
      "eval_steps_per_second": 16.106,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.1319090127944946,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5688,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6692233681678772,
      "eval_loss": 0.6350439190864563,
      "eval_matthews_correlation": 0.025622035134478298,
      "eval_runtime": 0.3205,
      "eval_samples_per_second": 3254.129,
      "eval_steps_per_second": 15.6,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.7428271770477295,
      "learning_rate": 4.5e-05,
      "loss": 0.5277,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6586769223213196,
      "eval_loss": 0.6556610465049744,
      "eval_matthews_correlation": 0.05555272752389579,
      "eval_runtime": 0.3151,
      "eval_samples_per_second": 3310.249,
      "eval_steps_per_second": 15.869,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.063387393951416,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.4781,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6586769223213196,
      "eval_loss": 0.7021281719207764,
      "eval_matthews_correlation": 0.06726048609499759,
      "eval_runtime": 0.3139,
      "eval_samples_per_second": 3322.475,
      "eval_steps_per_second": 15.927,
      "step": 204
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.0660338401794434,
      "learning_rate": 4.3e-05,
      "loss": 0.4279,
      "step": 238
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6490891575813293,
      "eval_loss": 0.7524747252464294,
      "eval_matthews_correlation": 0.06280376398791969,
      "eval_runtime": 0.3301,
      "eval_samples_per_second": 3159.185,
      "eval_steps_per_second": 15.145,
      "step": 238
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.1008501052856445,
      "learning_rate": 4.2e-05,
      "loss": 0.3883,
      "step": 272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6308724880218506,
      "eval_loss": 0.7883565425872803,
      "eval_matthews_correlation": 0.08342146557730178,
      "eval_runtime": 0.3069,
      "eval_samples_per_second": 3398.506,
      "eval_steps_per_second": 16.292,
      "step": 272
    },
    {
      "epoch": 8.0,
      "step": 272,
      "total_flos": 1793902531485696.0,
      "train_loss": 0.5260348284945768,
      "train_runtime": 45.4952,
      "train_samples_per_second": 9397.689,
      "train_steps_per_second": 37.367
    }
  ],
  "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
}