LongLLaMA is a large language model capable of handling long contexts. It is based on OpenLLaMA and fine-tuned with the Focused Transformer (FoT) method. - Aschen | Sharing the webz

Delete Set public Set private

1306 shaares

1306 shaares

Filters

Links per page

20 50 100

LongLLaMA is a large language model capable of handling long contexts. It is based on OpenLLaMA and fine-tuned with the Focused Transformer (FoT) method.

Un fork de LLaMA qui peut utiliser des contextes de 256K tokens contre 2048 pour le modèle originel.

Le code et le modèle entrainé sont sous licence Apache 2.0

c-ai · text-ai · open-source

July 9, 2023 at 21:57:46 UTC * · permalink

·

https://github.com/CStanKonrad/long_llama

Filters

Links per page

20 50 100