English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recent advances have greatly increased the capabilities of large languagemodels (LLMs), but our understanding of the models and their safety has notprogressed as fast. In this paper we aim to understand LLMs deeper by studyingtheir individual neurons. We build upon previous work showing large languagemodels such as GPT-4 can be useful in explaining what each neuron in a languagemodel does. Specifically, we analyze the effect of the prompt used to generateexplanations and show that reformatting the explanation prompt in a morenatural way can significantly improve neuron explanation quality and greatlyreduce computational cost. We demonstrate the effects of our new prompts inthree different ways, incorporating both automated and human evaluations.

The Importance of Prompt Tuning for Automated Neuron Explanations