Hugging Face's logo
BlinkDL's picture
38 2 32

BlinkDL

BlinkDL
batman9x's profile picture DoodleBear's profile picture pool-22's profile picture
·
https://www.rwkv.com/

AI & ML interests

RWKV is all you need

Recent Activity

liked a model about 1 hour ago
featherless-ai/Qwerky-72B-Preview
upvoted a paper 12 days ago
Training Language Models for Social Deduction with Multi-Agent Reinforcement Learning
updated a model 13 days ago
BlinkDL/rwkv-7-world
View all activity

Organizations

RWKV's profile picture FreedomAI's profile picture rwkv-x-dev's profile picture Social Post Explorers's profile picture

BlinkDL's activity

posted an update 2 months ago
view post
Post
3872
RWKV-7 "Goose" 0.4B trained w/ ctx4k automatically extrapolates to ctx32k+, and perfectly solves NIAH ctx16k 🤯 100% RNN and attention-free. Only trained on the Pile. No finetuning. Replicable training runs. tested by our community: https://github.com/Jellyfish042/LongMamba
posted an update 3 months ago
view post
Post
6336
RWKV-6-world-v3 (+3.1T tokens) is our best multilingual 7B model as of now: BlinkDL/rwkv-6-world

It's 100% RNN and attention-free. MMLU 54.2% (previous world-v2.1 = 47.9%. note: without eval-boosting tricks such as annealing).

RWKV-7-world-v4 soon :)
  • 1 reply
·
posted an update 5 months ago