註冊並分享邀請連結,可獲得影片播放與邀請獎勵。

Grigory Sapunov
@che_shr_cat
PhD in AI | GDE in AI | CTO Intento | Author "Deep Learning with JAX" 📝 ML insights: 🤖 AI reviews:
加入 June 2009
1.8K 正在關注    4.6K 粉絲
1/ From 7B to 30B parameters, zeroing fewer than 0.05% of frozen weights with GRPO matches full fine-tuning on math, and on Qwen2.5-7B it beats the base model even at pass@128 (80.08% vs 74.01%) without writing a single new weight. 🧵
顯示更多
0
7
112
9
轉發到社區