INDEX

Circuit Sparsity

No connected neurons data available

Explanations

element tree import (negative activations)

outer function returning inner function

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

arehouse

-0.04

udent

-0.04

light

-0.04

uild

-0.04

stac

-0.03

liver

-0.03

rontier

-0.03

 Tuple

-0.03

mployee

-0.03

ruck

-0.03

POSITIVE LOGITS

_i

0.03

Ã

0.03

bs

0.02

(v

0.02

fo

0.02

by

0.02

(a

0.02

dom

0.02

ef

0.01

ld

0.01

Activations Density 98.951%