Embeddings & Representations
The lookup table, the residual stream, contextual vectors, geometry and superposition.
11concepts
47flashcards
81minutes of reading
- 01 Embedding Geometry and Anisotropy Learned embedding spaces are not the well-spread sphere the geometric intuition suggests, they collapse into a narrow cone, and that fact quietly breaks naive similarity comparisons built on top of them.
- 02 Superposition and Polysemantic Neurons Individual neurons in a trained network routinely fire for several unrelated concepts at once, and the leading explanation is not noise, it is a model deliberately packing more features than it has dimensions using near-orthogonal directions.
- 03 The Residual Stream Reframing a transformer's residual connections as one shared, additive vector space that every layer reads from and writes to, the lens that makes attention, MLPs, and interpretability results legible as a single system.
- 04 Unembedding and the Logit Lens Multiplying an intermediate layer's residual stream by the final unembedding matrix, as if it were the last layer, turns out to produce surprisingly sensible next-token guesses, a cheap window into what a model has committed to mid-computation.