The Transformer Memory Architecture
Viewing transformers as long-term memory, working memory, and associative retrieval — and how Anthropic’s Workspace and J-Lens results fit into a memory-centric picture of the residual stream.
Viewing transformers as long-term memory, working memory, and associative retrieval — and how Anthropic’s Workspace and J-Lens results fit into a memory-centric picture of the residual stream.