LocalLLaMA

2878 readers

19 users here now

Welcome to LocalLLaMA! Here we discuss running and developing machine learning models at home. Lets explore cutting edge open source neural network technology together.

Get support from the community! Ask questions, share prompts, discuss benchmarks, get hyped at the latest and greatest model releases! Enjoy talking about our awesome hobby.

As ambassadors of the self-hosting machine learning community, we strive to support each other and share our enthusiasm in a positive constructive way.

founded 2 years ago

MODERATORS

chargoddard's frankensteined 22B llama2 (huggingface.co)

submitted 2 years ago by noneabove1182 to c/localllama

2 comments fedilink hide all child comments

This is Llama 2 13b with some additional attention heads from original-flavor Llama 33b frankensteined on.

Fine-tuned on ~10M tokens from RedPajama to settle in the transplants a little.

Not intended for use as-is - this model is meant to serve as a base for further tuning, hopefully with a greater capacity for learning than 13b.

top 2 comments

sorted by: hot top controversial new old

[–] [email protected] 3 points 2 years ago (1 children)

Some mad scientist action, I love it.

[–] noneabove1182 2 points 2 years ago

it apparently has larger implications than we might think, there was a big discussion in thebloke's discord