Experimental use
#144
by
yassineelkhadiri14
- opened
Hello There, I am using Mixtral-8x7B in a specific task and I was wondering if there any possibility to limit the model context windows to a fixed number of tokens.