metadata
license: apache-2.0

It's Edgerunners/meta-llama-3-8b-instruct-hf-ortho-baukit-10fail-1000total
But with extra ablations where the harmless dataset was OpusStories and the harmful dataset was llm-attacks.
license: apache-2.0
It's Edgerunners/meta-llama-3-8b-instruct-hf-ortho-baukit-10fail-1000total
But with extra ablations where the harmless dataset was OpusStories and the harmful dataset was llm-attacks.