RSSAmplifier

Convergent Thinking · Jan 10, 2026

AVnorm

0
Sign in to vote or save

This site does not allow itself to be embedded. You can still read it on the original site — the toolbar below keeps your place in the directory.

“Attention normalizes the wrong norm” showed that attention output variance collapses as context grows. L2-normalized softmax fixes it in theory, but existing architectures are tuned for L1. So, swapping the norm breaks hyperparameters. There’s a simpler fix: normalize the output. av = attn @ v # or F.scaled_dot_product_attention(...) av = GroupNorm(num_groups = heads,…

Read on convergentthinking.sh

Comments

Nothing yet. Say the first thing.

    Sign in to join the conversation.