Re:Weird GPT-4 behavior for the specific string “ davidjl”
by ntonozzi on Thursday June 08, 2023 @01:03PM
(#36245143)
The most effective architecture for LLMs without tokenization is Meta's MEGABYTE architecture: https://arxiv.org/abs/2305.07185. The sacrifice here is that they have two different types of layers, instead of the simple uniform transformer layers.
[ Parent
| on HN ]
There are no comments on this comment yet.