search

AI-Hypercomputer / maxtext

A simple, performant and scalable Jax LLM!
Apache License 2.0
1.54k stars 295 forks source link

Correct vocab size for 8x22b #1012

Closed RissyRan closed 3 weeks ago

RissyRan commented 3 weeks ago

Description

Correct vocab size from 32000 to 32768, which aligned with right model config