search

dusty-nv / jetson-containers

Machine Learning Containers for NVIDIA Jetson and JetPack-L4T
MIT License
1.9k stars 416 forks source link

mlc_llm_build fails for Mistral-7B-Instruct-v0.2 #453

Open rgobbel opened 3 months ago

rgobbel commented 3 months ago

Mistral-7B-Instruct-v0.2 does not use sliding window attention, so building fails because there is an assumption that Mistral models all have some value for sliding_window.