Towards Monosemanticity: Decomposing Language Models with Dictionary Learning (transformer-circuits.pub) | HN Screenshots

YHacker Newsscreenshots

Screenshot of transformer-circuits.pub: Towards Monosemanticity: Decomposing Language Models with Dictionary Learning

First screen, captured 3 years ago.

Looks similar

Screenshot of transformer-circuits.pub: Circuit Tracing: Revealing Computational Graphs in Language Models (Anthropic)

Circuit Tracing: Revealing Computational Graphs in Language Models (Anthropic)(transformer-circuits.pub)

161 points by ydnyshhh 1 year ago | 27 comments

Screenshot of transformer-circuits.pub: Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet

Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet (transformer-circuits.pub)

136 points by 1wheel 2 years ago | 26 comments

Screenshot of transformer-circuits.pub: Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet

Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet (transformer-circuits.pub)

19 points by Anon84 2 years ago | 1 comments

Screenshot of transformer-circuits.pub: When models manipulate manifolds: The geometry of a counting task

When models manipulate manifolds: The geometry of a counting task (transformer-circuits.pub)

84 points by vinhnx 1 year ago | 10 comments

Screenshot of transformer-circuits.pub: Emergent Introspective Awareness in Large Language Models

Emergent Introspective Awareness in Large Language Models (transformer-circuits.pub)

7 points by og_kalu 1 year ago | 1 comments

Screenshot of transformer-circuits.pub: A Mathematical Framework for Transformer Circuits (2021)

A Mathematical Framework for Transformer Circuits (2021)(transformer-circuits.pub)

95 points by Bluestein 1 month ago | 17 comments