← 返回卡片

原书出处

Build a Reasoning Model (From Scratch)

MEAP Version 5(早期访问版) · 核验版本的摘录快照 · 导入于 2026-11-14T08:12:58.963Z

loss values themselves should not be over-interpreted. Steps with reward_avg=0.000 produce a near-zero loss because all rollouts receive the same reward, resulting in vanishing group-relative advantages and little to no gradient signal. Larger loss magnitudes, such as at step 3, simply reflect bigger relative differences between rollouts and are typical for GRPO-style objectives, especially at the beginning of training.

此预览只包含定位所需的摘录和上下文。

原书定位详情

来源:a19c7933-4eda-4176-aebd-30fe7e1ad292

正文位置:extracted/Build a Reasoning Model (From Scratch)/_epub-src/OEBPS/Text/chapter-6.html

EPUB SHA-256:52d4d07986020be70f5aac21762dfcf60fcba08db0ba29d4f63957caf12aa377