Liberty

liberty4321's activity

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

commented on 🐯 Liger GRPO meets TRL 2 months ago

view reply

Since Liger kernel saves GPU memory. You can increase training throughput by increasing the batch size.

published an article 2 months ago

view article

Article

🐯 Liger GRPO meets TRL

By and 5 others •

May 25

• 48

New activity in diffusers/stable-diffusion-xl-1.0-inpainting-0.1 11 months ago

Update README.md

1

#24 opened 11 months ago by

liberty4321