Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Docs
  • Enterprise
  • Pricing

  • Log In
  • Sign Up

shisa-v2-dev

community
https://github.com/shisa-ai/shisa-v2
shisa-ai
Activity Feed

AI & ML interests

None defined yet.

Recent Activity

leonardlin  updated a Space 4 days ago
shisa-v2-dev/README
leonardlin  published a Space 4 days ago
shisa-v2-dev/README
View all activity

NekoMikoReimu's profile picture lhl's profile picture
Organization Card
Community About org cards

These are the archival development ablations for Shisa V2 family of Japanese multilingual LLMs.

This includes the work done on Llama 3.1 Shisa V2 405B, the strong Japanese model (open or closed) ever developed in Japan at the time of its release.

models 15

shisa-v2-dev/meti-geniac-405b-dpo3

406B • Updated Apr 27

shisa-v2-dev/meti-geniac-405b-dpo2

406B • Updated Apr 26

shisa-v2-dev/global_step6000_hf

406B • Updated Apr 24

shisa-v2-dev/global_step5500_hf

406B • Updated Apr 23

shisa-v2-dev/global_step5000_hf

406B • Updated Apr 22

shisa-v2-dev/global_step4500_hf

406B • Updated Apr 22

shisa-v2-dev/global_step4000_hf

406B • Updated Apr 20

shisa-v2-dev/global_step3500_hf

406B • Updated Apr 20

shisa-v2-dev/ablation-135-geniac.gbs128.2e6-shisa-v2-llama-3.1-8b

Text Generation • 8B • Updated Apr 3 • 2 • 1

shisa-v2-dev/ablation-134-geniac.gbs128.5e6-shisa-v2-llama-3.1-8b

Text Generation • 8B • Updated Apr 3 • 1
View 15 models

datasets 0

None public yet
Company
TOS Privacy About Jobs
Website
Models Datasets Spaces Pricing Docs