zhipeng ma

mamazi00's activity

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

New activity in haydn-jones/gemma-4-E4B-it-RL 4 months ago

Issue with Reinforcement Learning Fine-tuning on Gemma4

#1 opened 4 months ago by

mamazi00

New activity in google/gemma-4-31B-it 4 months ago

When training models from the gemma4 series using GRPO, an abnormally high grad norm was observed

#84 opened 4 months ago by

mamazi00