trainer_state.json

{
  "best_metric": 0.6569345593452454,
  "best_model_checkpoint": "bert_tiny_lda_20_v1_qqp/checkpoint-4266",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 11376,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.3419596254825592,
      "learning_rate": 0.00098,
      "loss": 0.6612,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": 0.6581591367721558,
      "eval_runtime": 12.6846,
      "eval_samples_per_second": 3187.326,
      "eval_steps_per_second": 12.456,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.10207448154687881,
      "learning_rate": 0.00096,
      "loss": 0.659,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": 0.6575562953948975,
      "eval_runtime": 12.7272,
      "eval_samples_per_second": 3176.652,
      "eval_steps_per_second": 12.414,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.05861695855855942,
      "learning_rate": 0.00094,
      "loss": 0.6587,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": 0.6569345593452454,
      "eval_runtime": 12.6759,
      "eval_samples_per_second": 3189.518,
      "eval_steps_per_second": 12.465,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.08356954157352448,
      "learning_rate": 0.00092,
      "loss": 0.6585,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": 0.6573521494865417,
      "eval_runtime": 12.632,
      "eval_samples_per_second": 3200.604,
      "eval_steps_per_second": 12.508,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.12464679032564163,
      "learning_rate": 0.0009000000000000001,
      "loss": 0.6585,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": 0.6573521494865417,
      "eval_runtime": 12.6421,
      "eval_samples_per_second": 3198.044,
      "eval_steps_per_second": 12.498,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.03694356605410576,
      "learning_rate": 0.00088,
      "loss": 0.6585,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": 0.658168613910675,
      "eval_runtime": 12.4501,
      "eval_samples_per_second": 3247.364,
      "eval_steps_per_second": 12.691,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.050063714385032654,
      "learning_rate": 0.00086,
      "loss": 0.6585,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": 0.6573427319526672,
      "eval_runtime": 12.5094,
      "eval_samples_per_second": 3231.962,
      "eval_steps_per_second": 12.63,
      "step": 9954
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.11186650395393372,
      "learning_rate": 0.00084,
      "loss": 0.6585,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": 0.6573427319526672,
      "eval_runtime": 12.4754,
      "eval_samples_per_second": 3240.777,
      "eval_steps_per_second": 12.665,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "step": 11376,
      "total_flos": 7.633075201391002e+16,
      "train_loss": 0.6589229814446258,
      "train_runtime": 1678.7379,
      "train_samples_per_second": 10836.891,
      "train_steps_per_second": 42.353
    }
  ],
  "logging_steps": 1,
  "max_steps": 71100,
  "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": 7.633075201391002e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}