Hugging Face's logo

LoRA-Learns-Less-and-Forgets-Less
community
https://arxiv.org/abs/2405.09673
Activity Feed

AI & ML interests

None defined yet.

Organization Card
Community About org cards

These are the model weights associated with the TMLR 2024 publication LoRA Learns Less and Forgets Less (Biderman et al. 2024). This work was done in collaboration with Databricks Mosaic AI Research.

Collections 4

Continued Pretraining - Code (StarCoder-Python)
This collection contains checkpoints and LoRA adapters for Llama-2-7B trained on the Python subset of StarCoder for up to 20 billion tokens.
Continued Pretraining - Code (StarCoder-Python)
This collection contains checkpoints and LoRA adapters for Llama-2-7B trained on the Python subset of StarCoder for up to 20 billion tokens.
View 4 collections

datasets 0

None public yet