experiments_deep_baseline/logs/config1_modernbert_large_seed42.json

{
  "run_id": "config1_modernbert_large_seed42",
  "config": {
    "name": "config1_modernbert_large",
    "owner": "Deep",
    "model_name": "answerdotai/ModernBERT-large",
    "max_length": 384,
    "learning_rate": 1.5e-05,
    "per_device_batch_size": 16,
    "grad_accumulation": 4,
    "epochs": 4,
    "weight_decay": 0.01,
    "warmup_ratio": 0.1,
    "loss_type": "ce",
    "focal_gamma": 2.0,
    "label_smoothing": 0.03,
    "use_lora": false
  },
  "seed": 42,
  "best_threshold": 0.475,
  "best_val_metrics": {
    "threshold": 0.475,
    "accuracy": 0.9990128331688055,
    "precision_injection": 1.0,
    "recall_injection": 0.996742671009772,
    "f1_injection": 0.99836867862969,
    "f1_macro": 0.9988304822730899,
    "pr_auc": 0.9999581050933732,
    "roc_auc": 0.9999815448782423,
    "fpr_safe": 0.0,
    "tp": 306,
    "tn": 706,
    "fp": 0,
    "fn": 1
  },
  "test_metrics": {
    "threshold": 0.475,
    "accuracy": 0.9990138067061144,
    "precision_injection": 0.9967532467532467,
    "recall_injection": 1.0,
    "f1_injection": 0.9983739837398374,
    "f1_macro": 0.9988331348281636,
    "pr_auc": 1.0,
    "roc_auc": 1.0,
    "fpr_safe": 0.0014144271570014145,
    "tp": 307,
    "tn": 706,
    "fp": 1,
    "fn": 0
  },
  "latency_ms_per_sample": 649.0914300084114,
  "train_time_sec": 133216.15459275246,
  "checkpoint_dir": "layer 2\\experiments_deep_baseline\\checkpoints\\config1_modernbert_large_seed42"
}