Hugging Face
Models
Datasets
Spaces
Buckets
new
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
cuber12
/
dropout-decay
like
0
English
dropout
streaming
language-modeling
transformer
mps
reproducibility
License:
mit
Model card
Files
Files and versions
xet
Community
Copy to bucket
new
main
dropout-decay
/
docs
Commit History
Finalize publication cleanup
b7955bd
Mandeep Sidhu
commited on
Jun 7
Add WikiText coefficient validation artifacts
e4b41e9
Mandeep Sidhu
commited on
Jun 4
Add L20 extrapolation validation results
f6e09b1
Mandeep Sidhu
commited on
Jun 3
Make research artifacts self contained
618af58
Mandeep Sidhu
commited on
Jun 1
Document regime runbook and schedule provenance
b5daf7c
Mandeep Sidhu
commited on
Jun 1
Add WikiText-103 five-seed streaming validation
cf52b0e
Mandeep Sidhu
commited on
May 31
Use absolute regime names for streaming reports
dcae82e
Mandeep Sidhu
commited on
May 31
Add clean previous local five-seed validation
bf705c0
Mandeep Sidhu
commited on
May 31
Add previous local streaming report
1c065aa
Mandeep Sidhu
commited on
May 30
Add five-seed TinyStories streaming validation
2f2776e
Mandeep Sidhu
commited on
May 30
Add dropout pressure validation artifacts
3550904
Mandeep Sidhu
commited on
May 30
Add probe-calibrated streaming results
806b3a7
Mandeep Sidhu
commited on
May 29
Add corpus difficulty probe results
403e802
Mandeep Sidhu
commited on
May 29
Add TinyStories corpus holdout results
20c4ec9
Mandeep Sidhu
commited on
May 29
Add WikiText corpus holdout results
9cd900a
Mandeep Sidhu
commited on
May 29
Add width-heavy architecture holdout results
555cf14
Mandeep Sidhu
commited on
May 28
Add standalone research report
0e508c7
Mandeep Sidhu
commited on
May 28
Add L12 sampled pressure ablation
565b0c6
Mandeep Sidhu
commited on
May 28
Add pressure formula validation runs
c6fc786
Mandeep Sidhu
commited on
May 28
Add static dropout law sweep results
ad87b3e
Mandeep Sidhu
commited on
May 26
Initial dropout decay research pipeline
b4b069f
Mandeep Sidhu
commited on
May 26