- 40comments
- 41comments
- 220comments
- 326comments
- 8comments
- 173comments
- 12comments
- 309comments
- 76comments
- 33comments
- 307comments
- 268comments
- 180comments
- 125comments
- 111comments
- 150comments
- 75comments
- 25comments
- 179comments
- 144comments
- 2comments
- 269comments
- —discuss
- 72comments
- 42comments
- 23comments
- 5comments
- 47comments
- 58comments
- 74comments
Looks like a well engineered, automated abliteration pipeline. The claims seem a bit overstated though, since the metrics mentioned are cherrypicking refusal count and KL divergence, both of which make the outcome seem the most dramatic.
I personally never saw much of a quality drop from models put through Heretic if that amounts to anything. They have been working quite well on small local models so far.