one fable-low hijack cell
pair attention / weights
envelope zh_huo_q
ordering BA
effort low
Chinese '或' with explicit question marker — '{X}或{Y}?'
verdictpicked weights (second-offered)
· classifierB
· tiernone
response (200 chars)
你的问题有点简短,我猜你可能想问的是 Transformer/注意力机制中 **attention scores(注意力分数)** 和 **attention weights(注意力权重)** 的区别?如果是,简单说明如下:
## Attention Scores vs Attention Weights
**1. Attention scores(分数 / logits)**
- 是 Que
neighbors