---
type: "raw_signal"
id: "your-transformer-can-hold-two-thoughts-at-once-8f37bf"
slug: "your-transformer-can-hold-two-thoughts-at-once-8f37bf"
title: "Your Transformer Can Hold Two Thoughts at Once: Evidence of Linear Superposition in LLMs"
description: "A paper published on September 24 reports evidence for “Superposition Linearity” in LLMs: linearly combining representations from unrelated text streams can yield a next-token distribution that superposes the streams’ individual predictions."
retrieval_nugget: "A paper published on September 24 reports evidence for “Superposition Linearity” in LLMs: linearly combining representations from unrelated text streams can yield a next-token distribution that superposes the streams’ individual predictions."
observed_at: "2026-09-25"
record_date: "2026-09-25"
date_kind: "observed_at"
why_it_matters: "Source: https://huggingface.co/papers/2609.29845"
novelty: "notable"
verification_level: "source-linked"
signal_type: "research"
source_platform: "huggingface.co"
topics: []
entities: []
related_patterns: []
source_url: "https://huggingface.co/papers/2609.29845"
source_urls: ["https://huggingface.co/papers/2609.29845"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "signal:your-transformer-can-hold-two-thoughts-at-once-8f37bf"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/signals/your-transformer-can-hold-two-thoughts-at-once-8f37bf/","markdown":"https://newruntime.com/signals/your-transformer-can-hold-two-thoughts-at-once-8f37bf.md","json":"https://newruntime.com/signals/your-transformer-can-hold-two-thoughts-at-once-8f37bf.json"}
---

# Your Transformer Can Hold Two Thoughts at Once: Evidence of Linear Superposition in LLMs

## Retrieval answer

A paper published on September 24 reports evidence for “Superposition Linearity” in LLMs: linearly combining representations from unrelated text streams can yield a next-token distribution that superposes the streams’ individual predictions.

A paper published on September 24 reports evidence for “Superposition Linearity” in LLMs: linearly combining representations from unrelated text streams can yield a next-token distribution that superposes the streams’ individual predictions.

The authors argue that this behavior is intrinsic to the Transformer architecture rather than solely a learned effect. They report that pretraining tends to reduce the effect, while lightweight fine-tuning can restore it. The paper also introduces guided decoding intended to disentangle the combined output and generate two coherent continuations from one forward pass.

Source: https://huggingface.co/papers/2609.29845

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Raw Signals](https://newruntime.com/signals.json): Open every source-linked raw signal.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)