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
chandar-lab
/
AMPLIFY_350M
Like
6
Follow
Chandar Research Lab
69
Safetensors
chandar-lab/UR100P
English
AMPLIFY
biology
custom_code
License:
mit
Model card
Files
Files and versions
xet
Community
Copy to bucket
new
main
AMPLIFY_350M
Commit History
May 28, 2025
18:54
Update rmsnorm.py
davidhd
Verified signature
223e35e
Modifies the forward pass of RMSNorm to avoid mixed precision issues as described in https://github.com/chandar-lab/AMPLIFY/issues/19
Feb 24, 2025
22:13
Fix attention mask dtype issues.
davidhd
Verified signature
35101e6
Fix typo from previous commit.
22:11
Add an assert about the attention_mask dtype.
davidhd
Verified signature
3fe920c
The attention mask in AMPLIFY should be additive, as a consequence, masked positions should have -inf values and unmasked positions should have 0 (or any constant value, but 0 is standard).
Jan 1, 2025
15:47
Fix typo in the learning rate of Stage 2
qfournier
Verified signature
f8b02da
Nov 4, 2024
19:50
Add support for CPU
qfournier
Verified signature
a735ae8
Oct 2, 2024
19:37
Update amplify.py
qfournier
Verified signature
858a98f
Oct 1, 2024
11:37
Update links
qfournier
Verified signature
50b926b
11:36
Update links
qfournier
Verified signature
f24f111
Sep 30, 2024
23:18
initial commit
qfournier
Verified signature
e726481
23:00
initial commit
qfournier
Verified signature
c7c5a31
