Thanks for sharing!!!
I might have spotted a minor mistake, should "To ensure full coverage of all domains in the non-thinking dataset" really be "To ensure full coverage of all domains in the thinking dataset"？

upvoted an article 24 days ago

Article

Why Maybe We're Measuring LLM Compression Wrong

•

Jun 21

• 9

liked a Space 5 months ago

2.87k

The Ultra-Scale Playbook

🌌

The ultimate guide to training LLM on large GPU Clusters

New activity in BAAI/Infinity-Instruct 11 months ago

About `system_prompt` setting when fine-tuning by this dataset

#22 opened 11 months ago by

Remixa

liked a model almost 2 years ago

xiaol/RWKV-v4-world-7B-one-state-65k

Updated Oct 12, 2023 • 4

liked a model over 2 years ago

stabilityai/stable-diffusion-2-1

Text-to-Image • Updated Jul 5, 2023 • 913k • 4.02k

RemixaAxun

AI & ML interests

Recent Activity

Organizations

Remixa's activity

Why Maybe We're Measuring LLM Compression Wrong

The Ultra-Scale Playbook

About `system_prompt` setting when fine-tuning by this dataset