INDEX

Circuit Sparsity

No connected neurons data available

Explanations

string literal content (negative activations)

year in datetime or range

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

mport

-0.03

 function

-0.02

dule

-0.02

import

-0.02

 import

-0.02

 True

-0.02

 usage

-0.02

'__

-0.01

visited

-0.01

 visited

-0.01

POSITIVE LOGITS

qrt

0.04

rior

0.03

lose

0.02

!=

0.02

uration

0.02

ault

0.02

ies

0.02

param

0.02

alue

0.02

args

0.02

Activations Density 86.808%