Kirill Shmilovich

KirillShmilovich's activity

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

New activity in SandboxAQ/AQAffinity 2 months ago

Clarification on two-head loss weighting vs inference output (head_2 upweighted, head_1 reported)

➕ 1

3

#15 opened 2 months ago by

KirillShmilovich

Questions for training / fine-tuning

➕ 3

4

#5 opened 3 months ago by

KirillShmilovich

New activity in kernels-community/vllm-flash-attn3 7 months ago

attention sinks & backward

4

#3 opened 9 months ago by

acforvs

Support for B200s?

👀 5

3

#7 opened 8 months ago by

shriramc