Glossary · Term

unembedding

← all terms

Definition

Plain language

The final step inside a language model that turns its internal state into actual word predictions.

As stated in the literature

The output projection matrix that maps a transformer's residual-stream representation back into vocabulary logits, often tied to the input embedding.

Why it matters: It is the bridge between the model's internal representations and the actual words it outputs, and a central object in interpretability work.

For example, the unembedding takes the model's final 4096-dimensional vector for a position and projects it to a score over all 100,000 vocabulary tokens.

Heard on the show

“Projections from middle layers through the unembedding can produce artifacts.”
Episode 094 — Chain-of-Thought Monitoring Fails Across Languages, and Worst Where It's Needed Most

Mentioned in 2 episodes

  1. 094
    Chain-of-Thought Monitoring Fails Across Languages, and Worst Where It's Needed Most
  2. 041
    When the Iteration Teaches the Model to Skip the Iteration

Related concepts

Related terms