raincandy-u

/

Llama-3-Aplite-Instruct-4x8B-MoE

like 38

Hi there, I was curious as to the best prompting method as the models that make up this one do not share the same prompting method, particularly the opus fine-tune which is a unique version of openchatml, so I was curious if there was a correct way to go about this, and how this potential problem was handled? Thanks