trainer_state.json

{
  "best_metric": 0.6573427319526672,
  "best_model_checkpoint": "bert_base_lda_5_v1_qqp/checkpoint-1422",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 8532,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.3263278603553772,
      "learning_rate": 0.00098,
      "loss": 0.6633,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": 0.6573427319526672,
      "eval_runtime": 25.7365,
      "eval_samples_per_second": 1570.918,
      "eval_steps_per_second": 6.139,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.06645578891038895,
      "learning_rate": 0.00096,
      "loss": 0.6589,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": 0.6579644680023193,
      "eval_runtime": 25.7118,
      "eval_samples_per_second": 1572.431,
      "eval_steps_per_second": 6.145,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.055242717266082764,
      "learning_rate": 0.00094,
      "loss": 0.6797,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6318327974276527,
      "eval_combined_score": 0.3159163987138264,
      "eval_f1": 0.0,
      "eval_loss": 0.658168613910675,
      "eval_runtime": 25.8276,
      "eval_samples_per_second": 1565.378,
      "eval_steps_per_second": 6.117,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.08355461061000824,
      "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": 25.6995,
      "eval_samples_per_second": 1573.185,
      "eval_steps_per_second": 6.148,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.12464568763971329,
      "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": 25.7443,
      "eval_samples_per_second": 1570.443,
      "eval_steps_per_second": 6.137,
      "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": 0.6579644680023193,
      "eval_runtime": 25.7207,
      "eval_samples_per_second": 1571.886,
      "eval_steps_per_second": 6.143,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "step": 8532,
      "total_flos": 2.8719571514554368e+17,
      "train_loss": 0.662904584681658,
      "train_runtime": 3658.7532,
      "train_samples_per_second": 4972.268,
      "train_steps_per_second": 19.433
    }
  ],
  "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": 2.8719571514554368e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}