INDEX

Circuit Sparsity

No connected neurons data available

Explanations

`print` function calls (negative activations)

less than or equal to condition

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 256 tokens each

Dataset (Dashboard)

neuronpedia/python-code-simplified

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 algorithm

-0.06

 Find

-0.05

 find

-0.05

dfs

-0.05

 iter

-0.05

reshold

-0.05

heck

-0.04

 steps

-0.04

dp

-0.04

_search

-0.04

POSITIVE LOGITS

if

0.05

==

0.05

0.03

If

0.03

0.02

.position

0.02

ld

0.02

(a

0.02

Activations Density 99.995%