Mistral 7B v0.2 Base (released at SHACK15sf hackathon)
Mistral 7B v0.2 Base (released at SHACK15sf hackathon)

github.com
GitHub - mistralai-sf24/hackathon

GitHub: https://github.com/mistralai-sf24/hackathon
\
X: https://twitter.com/MistralAILabs/status/1771670765521281370
New release: Mistral 7B v0.2 Base (Raw pretrained model used to train Mistral-7B-Instruct-v0.2)
\ ๐ธ https://models.mistralcdn.com/mistral-7b-v0-2/mistral-7B-v0.2.tar
\ ๐ธ 32k context window
\ ๐ธ Rope Theta = 1e6
\ ๐ธ No sliding window
\ ๐ธ How to fine-tune:
It does a little bit worse than v0.1 on all benchmarks which isn't ideal. That doesn't really say much about the finetuning potential though.