News for hackers. Stuff that matters.
[dead]
Re:Mixtral 8x7B: A sparse Mixture of Experts language model by Redster on Tuesday January 09, 2024 @12:01PM (#38928520)
At this small a scale, what would the benefits be of something like a 8x2B as opposed to moving to a 7B?
Re:Mixtral 8x7B: A sparse Mixture of Experts language model by lucidyan on Wednesday January 10, 2024 @06:21PM (#38945002)