← Back to article

Equation 4 · What a Circuit Explains: The State and Limits of Mechanistic Interpretability

What does this equation mean?

L(x)=∥x−x^(x)∥22+λ∥f(x)∥1,x^(x)=Wd f(x)+bd.\mathcal{L}(x) = \lVert x - \hat{x}(x) \rVert_2^2 + \lambda \lVert f(x) \rVert_1, \qquad \hat{x}(x) = W_d\, f(x) + b_d .

Read the formula alongside the article passage below. Each part has a deeper page with its role in the equation, the supporting passage and nearby citations.

This equation states an equality: the expressions on both sides have the same value under the article’s assumptions. Read the equation part by part below; each part has a contextual explanation and a link to its mathematical background.

Read it piece by piece

L\mathcal{L}

Symbol L

L is part of the quantity the equation computes from the expression on the right.

Understand this part →

xx

Symbol x

the writing.

Understand this part →

x^\hat{x}

Symbol hatx

hatx is one of the signed contributions combined to compute the quantity on the left.

Understand this part →

λ\lambda

Symbol λ

λ is one of the signed contributions combined to compute the quantity on the left.

Understand this part →

ff

Symbol f

f is one of the signed contributions combined to compute the quantity on the left.

Understand this part →

WdW_d

Symbol W_d

WdW_d is one of the signed contributions combined to compute the quantity on the left.

Understand this part →

bdb_d

Symbol b_d

bdb_d is one of the signed contributions combined to compute the quantity on the left.

Understand this part →

=

=

The expressions on both sides represent the same quantity under the stated assumptions.

Understand this part →

See an illustrated explanation →
addition

addition

Add the term after the plus sign to the term or group before it.

Understand this part →

subtraction

subtraction

Subtract the following term or group from the preceding one. A leading minus marks a negative quantity.

Understand this part →

subscript

subscript

The lower label selects a particular version, component, or indexed member of the quantity. For example, x₀ and xₜ can be values at different positions.

Understand this part →

superscript

superscript

A raised number can be a power. When it is a label or bound, it selects a case or the upper limit of a sum; the formula’s structure distinguishes these uses.

Understand this part →

See an illustrated explanation →

How to interpret it

Read it with the definitions, units, and assumptions supplied by the article.

What the article says around this equation

Now the careful part. Consider what the training objective actually asks for. Writing x for an activation vector, f(x) for the sparse code and x^\hat{x} for the reconstruction, the objective has the form L(x)=∥x−x^(x)∥22+λ∥f(x)∥1,x^(x)=Wd f(x)+bd\mathcal{L}(x) = \lVert x - \hat{x}(x) \rVert_2^2 + \lambda \lVert f(x) \rVert_1, \qquad \hat{x}(x) = W_d\, f(x) + b_d . Every term refers to the activation vector. No term refers to what the model does with that activation afterwards. The objective rewards a code that reconstructs the activation sparsely; it is indifferent to whether the dictionary elements correspond to anything the network’s downstream layers treat as a unit. Low reconstruction error at high sparsity is therefore evidence that the activation distribution is sparsely decomposable in the trained basis. It is not evidence…
Read the full surrounding passage
Now the careful part. Consider what the training objective actually asks for. Writing x for an activation vector, f(x) for the sparse code and x^\hat{x} for the reconstruction, the objective has the form L(x)=∥x−x^(x)∥22+λ∥f(x)∥1,x^(x)=Wd f(x)+bd\mathcal{L}(x) = \lVert x - \hat{x}(x) \rVert_2^2 + \lambda \lVert f(x) \rVert_1, \qquad \hat{x}(x) = W_d\, f(x) + b_d . Every term refers to the activation vector. No term refers to what the model does with that activation afterwards. The objective rewards a code that reconstructs the activation sparsely; it is indifferent to whether the dictionary elements correspond to anything the network’s downstream layers treat as a unit. Low reconstruction error at high sparsity is therefore evidence that the activation distribution is sparsely decomposable in the trained basis. It is not evidence that the basis is the model’s own.

Read the equation in its article →

Sources cited in the article section

These citations give research context. Read each source to check which claims it supports.

Return to What a Circuit Explains: The State and Limits of Mechanistic Interpretability

See this formula across 2 published contexts →

Browse the mathematical compendium →