Home Projects laser
laser
Python

laser

The Truth Is In There: Improving Reasoning in Language Models with Layer-Selective Rank Reduction

by pratyushasharma · GitHub
Stars
Forks
License
Created
Last commit
Category
Language
Likely
Self-hostable
gpt-jinterpretabilitylaserMITPythonSelf-hostable
View on GitHub
In plain words

Improve the reasoning abilities of language models by adjusting their internal layers for better performance.

You maintain this project?

Claim its page: indexed whatever its rank, translated into six languages, and enriched with what you write yourself.

Claim this page →
laser — GitHub preview card
📈 Star history
396395
2026-07-202026-08-31
📈 Track laser

Get an email alert on its next release or when it starts trending — never miss the moment.

Free · no card · unsubscribe anytime
Get email alerts →
📄 About

The Truth Is In There: Improving Reasoning in Language Models with Layer-Selective Rank Reduction

laser has 395 stars on GitHub. It has been forked 35 times. laser is written mainly in Python. It has been in active development since 2023. laser is available under the MIT license. Its main topics are gpt-j, interpretability, laser, llama2.

Frequently asked questions

What is laser?

The Truth Is In There: Improving Reasoning in Language Models with Layer-Selective Rank Reduction

Is laser open source?

laser is an open-source project. It is released under the MIT license.

Is laser free?

Yes. laser is free and open source — you can use, modify and self-host it.

What license does laser use?

laser is available under the MIT license.

What language is laser written in?

laser is written mainly in Python.

🏅 Maintainer of this project?
olud.ai badge — laser

Add this live badge to your README — your GitHub stars and directory rank, refreshed daily.

[![olud.ai](https://olud.ai/badge.php?tool=pratyushasharma-laser)](https://olud.ai/project/pratyushasharma-laser.html)
More badge options →
🧬 Shares DNA with🧬 View the DNA map →

Measured from GitHub topics shared by both projects, weighted by how rare each topic is.