← All parts of this equation

Equation 5 · Part 4 · The Hardest Unsolved Problems in Mechanistic Interpretability

Symbol g_j

di(m2)≈∑j=1m1gj dj(m1),∥g∥0≪m1,m1<m2,d_i^{(m_2)} \approx \sum_{j=1}^{m_1} g_j\, d_j^{(m_1)}, \qquad \lVert g \rVert_0 \ll m_1, \quad m_1 < m_2,
gjg_j

What this part means

gjg_j is an input to the expression that computes the quantity on the left.

Its job in the formula

gjg_j is an input to the expression that computes the quantity on the left.

The passage around this formula

Leask, Nanda and colleagues then tested the assumption directly with two new techniques, and the result undercuts the idea that there is a “right” dictionary size waiting to be found by scaling further. Stitching the dictionaries of differently sized sparse autoencoders together, they show that larger dictionaries recover latents genuinely missing from smaller ones — the smaller dictionary is incomplete. Training a second, “meta” sparse autoencoder on the decoder directions of a first one, they show that what looks like a single, atomic feature in a large dictionary is itself well approximated by a sparse combination of directions from a smaller one — a reported example decomposes an…

Read this part in the article →

Learn the underlying idea

A subscript is a label attached below a symbol. It often selects a time step, component, category, or member of a sequence.

Open the illustrated subscripts: which member of a family? guide →

See this notation across published equations →

Sources cited in the surrounding passage

These citations provide research context; check each source for the exact claim it supports.