Goodfire
@GoodfireAI
Using interpretability to understand, learn from, and design AI.
Joined August 2024
44 Following 27.4K Followers
Our team spent months developing RLFR, our method which uses probes on a model's internals as reward signals for RL.
Silico reproduced it in 2 days, reducing hallucinations in Qwen3-8B by 37% without capability loss. (3/6)
Show more
