๊ฐ€์ž… ํ›„ ์ดˆ๋Œ€ ๋งํฌ๋ฅผ ๊ณต์œ ํ•˜๋ฉด ๋™์˜์ƒ ์žฌ์ƒ ๋ฐ ์ดˆ๋Œ€ ๋ณด์ƒ์„ ๋ฐ›์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.

Grigory Sapunov
@che_shr_cat
PhD in AI | GDE in AI | CTO Intento | Author "Deep Learning with JAX" ๐Ÿ“ ML insights: ๐Ÿค– AI reviews:
๊ฐ€์ž… June 2009
1.8K ํŒ”๋กœ์ž‰ ์ค‘    4.6K ํŒฌ
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. ๐Ÿงต
๋” ๋ณด๊ธฐ