Hugging Face's logo

Test ORG

non-profit
Activity Feed

AI & ML interests

None defined yet.

Recent Activity

vicgalle  authored a paper 7 months ago
Merging Improves Self-Critique Against Jailbreak Attacks
vicgalle  authored a paper 9 months ago
Configurable Safety Tuning of Language Models with Synthetic Preference Data
vicgalle  updated a model 10 months ago
vicgalleorg/TruthfulQwen1.5-4B
View all activity

vicgalleorg's activity

vicgalle 
authored a paper 7 months ago
vicgalle 
authored a paper 9 months ago
vicgalle 
updated 2 models 10 months ago
vicgalle 
authored a paper 11 months ago
vicgalle 
posted an update 11 months ago
view post
Post
Can you merge models of different sizes? ⚗️

Well, yes, if the models are somewhat compatible. Here is an experiment I did. I wanted to merge two of the best performing models: mlabonne/NeuralBeagle14-7B and jeonsworld/CarbonVillain-en-10.7B-v4

Here is my recipe:
1. Expand the layers of NeuralBeagle to 10.7B ala frankenmerge.
2. DPO-tune the previous model with a high-quality preference dataset, argilla/distilabel-intel-orca-dpo-pairs
3. Merge the previous model with CarbonVillain (needs —allow-crimes in mergekit! 🔪)

And here is the resulting model, CarbonBeagle-11B, which ranked top in the leaderboard for its size class:
vicgalle/CarbonBeagle-11B
  • 2 replies
·
vicgalle 
authored a paper about 1 year ago
vicgalle 
authored 3 papers over 1 year ago