Pith Number

pith:66QXZ4FJ

pith:2025:66QXZ4FJYTGD4ZNQASST5IPCR7

not attested not anchored not stored refs pending

Interpreting the Repeated Token Phenomenon in Large Language Models

Federico Barbero, Ilia Shumailov, Itay Yona, Jamie Hayes, Yossi Gandelsman

arxiv:2503.08908 v1 · 2025-03-11 · cs.LG · cs.AI · cs.CL · cs.CR

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{66QXZ4FJYTGD4ZNQASST5IPCR7}

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

5 papers in Pith

Contribution Weights: A Geometrical Analysis of Self-Attention Transformers

Registers Matter for Pixel-Space Diffusion Transformers

Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free

The Structural Origin of Attention Sink: Variance Discrepancy, Super Neurons, and Dimension Disparity

Taming Outlier Tokens in Diffusion Transformers

Receipt and verification

First computed	2026-07-05T10:29:43.495974Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

f7a17cf0a9c4cc3e65b004a53ea1e28fecee601fca03d6d71a7428372971a6cb

Aliases

arxiv: 2503.08908 · arxiv_version: 2503.08908v1 · doi: 10.48550/arxiv.2503.08908 · pith_short_12: 66QXZ4FJYTGD · pith_short_16: 66QXZ4FJYTGD4ZNQ · pith_short_8: 66QXZ4FJ

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/66QXZ4FJYTGD4ZNQASST5IPCR7 \
  | 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: f7a17cf0a9c4cc3e65b004a53ea1e28fecee601fca03d6d71a7428372971a6cb

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "968768bbe6373d20e73b80e66504078875a98df2e0a4661c8af717f8317a524d",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.CL",
      "cs.CR"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2025-03-11T21:40:58Z",
    "title_canon_sha256": "3673431520ce2132b67d497789061d67ea330a1426670ea618def70c0a55f73b"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2503.08908",
    "kind": "arxiv",
    "version": 1
  }
}