Pith Number

pith:CH2BG5A2

pith:2024:CH2BG5A2FQN6GR74Z3AE3IMZRL

not attested not anchored not stored refs pending

Steering Language Model Refusal with Sparse Autoencoders

Blake Bullwinkel, David Majercak, Dean Carignan, Eric Horvitz, Forough Poursabzi-Sangdeh, Harsha Nori, Jingya Chen, Kyle O'Brien, Richard Edgar, Xavier Fernandes

arxiv:2411.11296 v2 · 2024-11-18 · cs.LG

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{CH2BG5A2FQN6GR74Z3AE3IMZRL}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Cited by

15 papers in Pith

Forecasting With LLMs: Improved Generalization Through Feature Steering

Pre-Intervention Prediction of Sparse Autoencoder Steering Side Effects

Perplexity Can Miss SAE Feature Damage Under Quantization

Palette: A Modular, Controllable, and Efficient Framework for On-demand Authorized Safety Alignment Relaxation in LLMs

At the Edge of Understanding: Sparse Autoencoders Trace The Limits of Transformer Generalization

Receipt and verification

First computed	2026-07-05T11:35:30.823505Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

11f413741a2c1be347fccec04da1998ac62b49c8d77c05a13fecee0843565416

Aliases

arxiv: 2411.11296 · arxiv_version: 2411.11296v2 · doi: 10.48550/arxiv.2411.11296 · pith_short_12: CH2BG5A2FQN6 · pith_short_16: CH2BG5A2FQN6GR74 · pith_short_8: CH2BG5A2

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/CH2BG5A2FQN6GR74Z3AE3IMZRL \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 11f413741a2c1be347fccec04da1998ac62b49c8d77c05a13fecee0843565416

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "ce59db6e76956b420d65cef75a44c38d490b7ed6d91ede8204652699cbfcb5fc",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2024-11-18T05:47:02Z",
    "title_canon_sha256": "f6a5f1fa0958319e821d1d864492b8e2ee8d74fe924f148c61effc00405b4435"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2411.11296",
    "kind": "arxiv",
    "version": 2
  }
}