MaximeRobeyns/bayesian_lora

Bayesian Low-Rank Adaptation for Large Language Models

45
/ 100
Emerging

This is a specialized library for researchers and practitioners working with Large Language Models (LLMs) that use Low-Rank Adaptation (LoRA). It helps you calculate important Bayesian quantities like Kronecker factors, marginal likelihood, and the posterior predictive distribution. You provide your LoRA-adapted LLM and data, and it outputs these statistical insights, enabling more robust model analysis and development.

Used by 1 other package. No commits in the last 6 months. Available on PyPI.

Use this if you are a machine learning researcher or engineer looking to apply Bayesian methods to understand the uncertainty and evidence of your LoRA-adapted LLMs.

Not ideal if you are looking for a general-purpose LLM fine-tuning tool or if you are not familiar with Bayesian statistics and LoRA.

large-language-models bayesian-statistics model-analysis machine-learning-research model-uncertainty
Stale 6m
Maintenance 0 / 25
Adoption 8 / 25
Maturity 25 / 25
Community 12 / 25

How are scores calculated?

Stars

37

Forks

5

Language

Python

License

Apache-2.0

Category

llm-fine-tuning

Last pushed

Jun 22, 2024

Commits (30d)

0

Dependencies

3

Reverse dependents

1

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/MaximeRobeyns/bayesian_lora"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.