laiyukyu's picture

4

laiyukyu

timmylai

·

AI & ML interests

None yet

Recent Activity

reacted to grimjim's post with 👍 15 days ago

I'm (finally) releasing a Python script that trims excess weights in Gemma2 full-weight models that bloated by ~1B parameters due to an early mergekit bug. https://github.com/jim-plus/Gemma2-mergekit-remediation I'd noticed something was off when merges of Gemma2 9B models ended up having ~10B parameters. The current mergekit package is fine, but there are still bloated models on HF that could stand to be fixed. The script assumes that it will be run from the same directory as the model weights, and will trim the unnecessary lm_head.weight tensor and corresponding index entry.

reacted to nroggendorff's post with ➕ 16 days ago

hey nvidia, can you send me a gpu? comment or react if you want ~~me~~ to get one too. 👉👈

reacted to danielhanchen's post with 👍 about 1 month ago

I uploaded GGUFs, 4bit bitsandbytes and full 16bit precision weights for Llama 3.3 70B Instruct are here: https://huggingface.co/collections/unsloth/llama-33-all-versions-67535d7d994794b9d7cf5e9f You can also finetune Llama 3.3 70B in under 48GB of VRAM with Unsloth! GGUFs: https://huggingface.co/unsloth/Llama-3.3-70B-Instruct-GGUF BnB 4bit: https://huggingface.co/unsloth/Llama-3.3-70B-Instruct-bnb-4bit 16bit: https://huggingface.co/unsloth/Llama-3.3-70B-Instruct

View all activity

Organizations

None yet

Collections 1

spaces 2

Argilla Space

Mattshumer Reflection Llama 3.1 70B

models

None public yet

datasets

None public yet