causal

Chain-of-thought transmits info via KV cache

Chain-of-thought doesn't just transmit the single output token; during the forward pass the model creates KV (key-value) values that future steps attend to, so the model can smush information about potential futures into the keys and values it might want, transmitting far more than the tiny log-vocab-size bits of one token.

causalpending

Speaker

Sholto Douglas

Evidence Quote

“it's learning to smush information about potential futures into the keys and values that it might want to use in order to predict future information.”

Source

Sholto Douglas & Trenton Bricken — How LLMs actually think— Dwarkesh Patel Podcast
Created: 6/13/2026, 3:31:11 AM

Your notes

Loading notes...