MillionthOdin16

3 ms·

1. 1.▲
   UMAF: Adapting LLMs Without Fine-Tuning(github.com)

   2 points by MillionthOdin16 2y ago|1 comments
2. 2.▲
   by MillionthOdin16 2y ago

   I'm experimenting with an approach called Universal Model Adapter Fusion to transfer capabilities LLMs without fine-tuning or distillation. The idea's to pull behaviors - like task-specific smarts or reasoning styles -from one mo
3. 3.▲
   by MillionthOdin16 4y ago

   You can run a demo with docker pretty quickly, but I didn't see any live demos. I was able to get it running with a one-line command though.