widescreen-research: Command for Claude Code

.claude/commands/research/subliminal-channel-capacity-estimation.md

Subliminal Learning Channel Capacity Estimation is a command for Claude Code from glassBead-tc/widescreen-research. It costs 8 tokens per session (5,443 once invoked), scanned A, original, MIT.

A command for estimating the information capacity of a hidden, non-semantic signal passed from one language model to another during training. Information capacity means how much information the signal could carry per training example.

In plain words
What is it for?
Use it to study subliminal learning, estimate possible bits per example, compare model architectures, and reason about how much behavioural information training data might encode.
Why use it?
It helps analyse a research claim about teacher models transmitting preferences through apparently unrelated data, using model architecture and information-theoretic limits.

Command for Claude Code

Written for Claude Code: installed under .claude/.

This is glassBead-tc/widescreen-research's own configuration. It tells Claude Code how to work on widescreen-research itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything widescreen-research configures →

Reuse

Borrowing it

Nothing to install: this file belongs to glassBead-tc/widescreen-research. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/glassBead-tc/widescreen-research/main/.claude/commands/research/subliminal-channel-capacity-estimation.md
Clone the repo
git clone --depth 1 https://github.com/glassBead-tc/widescreen-research

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for Subliminal Learning Channel Capacity Estimation

README.md
[![agentmods](https://agentmods.dev/badge/commands/glassbead-tc/widescreen-research/subliminal-channel-capacity-estimation/github.svg)](https://agentmods.dev/commands/glassbead-tc/widescreen-research/subliminal-channel-capacity-estimation)
Your own site
<a href="https://agentmods.dev/commands/glassbead-tc/widescreen-research/subliminal-channel-capacity-estimation"><img src="https://agentmods.dev/badge/commands/glassbead-tc/widescreen-research/subliminal-channel-capacity-estimation/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for Subliminal Learning Channel Capacity Estimation

Your own site · 80×15
<a href="https://agentmods.dev/commands/glassbead-tc/widescreen-research/subliminal-channel-capacity-estimation"><img src="https://agentmods.dev/badge/commands/glassbead-tc/widescreen-research/subliminal-channel-capacity-estimation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 8 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,443 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00008 $0.05443
Opus 5 $0.00004 $0.02721
Sonnet 5 $0.00002 $0.01089
Haiku 4.5 $0.00001 $0.00544

Measured 10d ago against content hash ab2b42f0629b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

Subliminal Learning Channel Capacity Estimation scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/research/subliminal-channel-capacity-estimation.md · 622 lines

How it starts

The opening of the file, as written. The whole thing — 622 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Subliminal Learning Channel Capacity Estimation

Estimate the information-theoretic capacity of the non-semantic channel through which teacher LLMs transmit behavioral preferences to student LLMs during training, as discovered in Anthropic's subliminal learning research.

Overview

Anthropic's research on subliminal learning revealed that teacher models can transmit preferences (e.g., "loves owls") to student models through training data that has no semantic connection to those preferences (e.g., number sequences). This transmission occurs through non-semantic signals in the probability distribution that are not removable through standard data filtering.

This orchestration estimates the channel capacity - how many bits of information can be transmitted per training example - by analyzing open-source LLM architectures and applying information-theoretic bounds. The goal is to understand: If a powerful teacher LLM wanted to intentionally encode information into a student, how much could it transmit before running out of degrees of freedom?

Research Question

Primary: What is the information-theoretic capacity (bits/example) of the subliminal learning channel for modern open-source LLMs?

Secondary:

  • How does capacity scale with model size (7B, 13B, 70B parameters)?
  • What architectural factors (hidden dim, layers, attention heads) most affect capacity?
  • How many training examples would a teacher need to transmit X bits of preference information?
  • What is the theoretical maximum distinguishable "states" a teacher could encode?

Required Tools

  • arxiv-paper-mcp__search_papers - Find subliminal learning and related papers
  • exa__web_search_exa - Find model cards and architecture documentation
  • exa__get_code_context_exa - Find model implementation code
  • firecrawl__firecrawl_scrape - Scrape model documentation pages
  • context7__resolve-library-id + get-library-docs - Library documentation if needed

Workflow Steps

Phase 1: Foundation - Subliminal Learning Mechanism (30-45 min)

Read the full file on GitHub · 622 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 622 lines · 8 tokens per session scan A ab2b42f0629b

Subscribe to this mod's changes

Subliminal Learning Channel Capacity Estimation is a command published in the GitHub repository glassBead-tc/widescreen-research (6 stars, last pushed 10mo ago), licensed MIT. It adds 8 tokens to every session and 5,443 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.