MNoorFawi/curlora
The code repository for the CURLoRA research paper. Stable LLM continual fine-tuning and catastrophic forgetting mitigation.
CURLoRA helps you continually update large language models (LLMs) with new information without making them "forget" what they already know. It takes an existing LLM and training data for new tasks, producing a fine-tuned model that remembers its prior knowledge while learning new capabilities. This is for machine learning engineers and researchers who need to adapt LLMs efficiently and reliably over time.
No commits in the last 6 months.
Use this if you need to fine-tune an LLM on new datasets sequentially and want to prevent the model from losing previously learned knowledge (catastrophic forgetting), especially with limited data.
Not ideal if you are looking for a plug-and-play solution for quantized models or if you are not comfortable modifying Python code to integrate this technique.
Stars
53
Forks
5
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Aug 28, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/MNoorFawi/curlora"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
OptimalScale/LMFlow
An Extensible Toolkit for Finetuning and Inference of Large Foundation Models. Large Models for All.
adithya-s-k/AI-Engineering.academy
Mastering Applied AI, One Concept at a Time
jax-ml/jax-llm-examples
Minimal yet performant LLM examples in pure JAX
young-geng/scalax
A simple library for scaling up JAX programs
riyanshibohra/TuneKit
Upload your data → Get a fine-tuned SLM. Free.