Content
@
0 reply
0 recast
0 reaction
Brian Kim
@brianjckim
how significant is this mitral release? cc @giu
2 replies
0 recast
0 reaction
ππͺπΎπ‘π‘πΎ
@gm8xx8
The emergence of this MoE model could mirror LLaMAβs impact on dense transformers, potentially uniting the open source community around a standardized MoE framework for collaborative innovation. Itβs pretty significant. Already some interesting buzz.
0 reply
0 recast
1 reaction
Giuliano Giacaglia π²
@giu
Probably the best model out there. It is a Mixture of experts but their goal was to beat ChatGPT 3.5 by a wide margin. I would assume it does. In terms of applications, it is v costly to run. I think it is likely that the models (distilling, etc) that come out of it become important
1 reply
0 recast
1 reaction
