lora/training_config.json

{
  "args": {
    "base": "Qwen/Qwen3.5-9B-Base",
    "n_per_source": 1000,
    "epochs": 2,
    "lr": 3e-05,
    "head_lr": 0.0,
    "weight_decay": 0.01,
    "lora": 16,
    "accum": 8,
    "holdout": "",
    "perm_kl": 0.0,
    "perm_frac": 0.3,
    "ord_w": 0.0,
    "label_smoothing": 0.0,
    "brier_w": 0.0,
    "focal_gamma": 0.0,
    "suite": null,
    "train_sources": "",
    "device": "mps",
    "batch": 1,
    "dtype": "fp32",
    "weights_dtype": "bf16",
    "checkpointing": 0,
    "option_isolation": 0,
    "special_embeddings": 0,
    "head_dim": 256,
    "lora_targets": "all",
    "base_revision": "",
    "p_none": 0.1,
    "p_none_distract": 0.12,
    "p_distract": 0.15,
    "p_none_pair": 0.0,
    "synthetic_repeat": 1,
    "public_frac": 1.0,
    "anchor": "",
    "anchor_w": 0.0,
    "anchor_sources": "",
    "out": "runs/novaeon-sentinel-9b",
    "data": "data/train-v3-2q.jsonl",
    "max_state": 704,
    "replay": 0,
    "init_from": "jaredpalmer/kev-9b",
    "seed": 1
  },
  "suite_sha256": null,
  "base_revision": null,
  "init_source": {
    "init_from": "jaredpalmer/kev-9b",
    "resolved": "jaredpalmer/kev-9b@2629c06a5aeb0feb3b9783bafed17ed8f39ecf5c",
    "adapter_sha256": "c02d0d1a70367b5779deb23b28153284b59c067038323cc17ddab5b92dfce72f",
    "head_sha256": "5528f777437f561cb4a9d56b0278a42b6ae6edd373e7921e52b13adaac988c9f",
    "adapter_tensors": 496
  },
  "ordinal_objective": "ranked_probability_score",
  "holdout": []
}