Mahesh Deshwal

deshwalmahesh's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

updated a model 1 day ago

deshwalmahesh/domyn-edge-jev-decision

Text Generation • 4B • Updated about 21 hours ago • 1.93k

New activity in domyn/Domyn-Edge-SFT 3 days ago

RoPE fp32 + gradient checkpointing + SDPA attention backend

1

#1 opened 3 days ago by

deshwalmahesh

published a model 3 days ago

deshwalmahesh/domyn-edge-jev-decision

Text Generation • 4B • Updated about 21 hours ago • 1.93k

upvoted an article 12 months ago

view article

Article

Continuous batching from first principles

• 
• 
• +1

ror, ArthurZ, mcpotato

•

Nov 25, 2025

• 462

liked a model over 2 years ago

facebook/musicgen-small

Text-to-Audio • 0.6B • Updated Nov 17, 2023 • 149k • 531

New activity in microsoft/Phi-3-mini-4k-instruct over 2 years ago

Does it support Flash Attention-2?

4

#12 opened over 2 years ago by

deshwalmahesh

commented a paper over 2 years ago

Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

Paper • 2404.14219 • Published Apr 22, 2024 • 262 •

42

New activity in microsoft/phi-2 almost 3 years ago

Training model breaks with flash_attn_2. Error "NameError: name 'index_first_axis' is not defined"

👍 1

12

#105 opened almost 3 years ago by

praveeny

liked a model over 3 years ago

Lykon/DreamShaper

Text-to-Image • Updated Apr 14, 2024 • 364k • • 1.03k