trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6237125992774963,
  "best_model_checkpoint": "bert_base_km_100_v1_cola/checkpoint-34",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.0520387887954712,
      "learning_rate": 4.9e-05,
      "loss": 0.6167,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6237125992774963,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.6698,
      "eval_samples_per_second": 1557.141,
      "eval_steps_per_second": 7.465,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6833245754241943,
      "learning_rate": 4.8e-05,
      "loss": 0.5951,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6893576383590698,
      "eval_loss": 0.6252963542938232,
      "eval_matthews_correlation": 0.012612883800688155,
      "eval_runtime": 0.6909,
      "eval_samples_per_second": 1509.517,
      "eval_steps_per_second": 7.236,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.7721410989761353,
      "learning_rate": 4.7e-05,
      "loss": 0.5382,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6711409687995911,
      "eval_loss": 0.6292206048965454,
      "eval_matthews_correlation": 0.09996873695642235,
      "eval_runtime": 0.6728,
      "eval_samples_per_second": 1550.343,
      "eval_steps_per_second": 7.432,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.8792331218719482,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.456,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6490891575813293,
      "eval_loss": 0.6964066028594971,
      "eval_matthews_correlation": 0.13304398041338622,
      "eval_runtime": 0.6917,
      "eval_samples_per_second": 1507.821,
      "eval_steps_per_second": 7.228,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.7057266235351562,
      "learning_rate": 4.5e-05,
      "loss": 0.3717,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6299136877059937,
      "eval_loss": 0.7489902973175049,
      "eval_matthews_correlation": 0.06518120581873378,
      "eval_runtime": 0.6744,
      "eval_samples_per_second": 1546.528,
      "eval_steps_per_second": 7.414,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.225237846374512,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.298,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6423777341842651,
      "eval_loss": 0.9227033853530884,
      "eval_matthews_correlation": 0.07676746462839329,
      "eval_runtime": 0.6782,
      "eval_samples_per_second": 1537.831,
      "eval_steps_per_second": 7.372,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 6749587903150080.0,
      "train_loss": 0.4792768253999598,
      "train_runtime": 100.6716,
      "train_samples_per_second": 4246.979,
      "train_steps_per_second": 16.887
    }
  ],
  "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": 6749587903150080.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}