kuleshov-group / llmtools

Finetuning Large Language Models on One Consumer GPU in Under 4 Bits

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Larger Context Windows

griff4692 opened this issue · comments

Hi -

Do you, or (can you!), offer support for context windows > 2048?

I'd love to be able to fine-tune one of the longer range models (e.g., Mosaic).

I have long inputs and need around 4k tokens.

Thanks!