Hugging Face
Search models, datasets, users...
Ctrl K
Models
Datasets
Spaces
Buckets
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
mX7pf <img src=x onerror=window.__pf1=1>
jdjsjsj28828
1
Follow
jdjdje8383's profile picture
Chris4K's profile picture
2 followers
ยท
3 following
https://example.com/
AI & ML interests
None yet
Recent Activity
reacted
to
Chris4K
's
post
with ๐
about 5 hours ago
๐ง Zero-FLOP Memory Can a small transformer do more by remembering differently? 39.5M parameters + discrete count tables + document cache. 1.017 BPC on enwik8. No gradient updates needed to build the memory. Exploring a different way to scale language modeling: not just more parameters, but better memory. Paper & code follows ... if someones interested
replied
to
Chris4K
's
post
about 5 hours ago
๐ง Zero-FLOP Memory Can a small transformer do more by remembering differently? 39.5M parameters + discrete count tables + document cache. 1.017 BPC on enwik8. No gradient updates needed to build the memory. Exploring a different way to scale language modeling: not just more parameters, but better memory. Paper & code follows ... if someones interested
published
a bucket
about 14 hours ago
jdjsjsj28828/bkprobe
View all activity
Organizations
jdjsjsj28828
's models
2
Sort:ย Recently updated
jdjsjsj28828/mx7n
Updated
about 14 hours ago
jdjsjsj28828/tcprobe
Token Classification
โข
Updated
about 14 hours ago