INDEX

Circuit Sparsity

No connected neurons data available

Explanations

comma separating arguments (negative activations)

variable assignment or sum calculation

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

etween

-0.06

dfs

-0.05

_with

-0.05

ith

-0.05

abs

-0.05

ta

-0.04

whe

-0.04

 math

-0.04

gorithm

-0.04

 algorithm

-0.04

POSITIVE LOGITS

plt

0.06

alculate

0.05

zero

0.04

cond

0.04

 remaining

0.04

 calculate

0.04

.add

0.04

ange

0.04

 last

0.03

ameters

0.03

Activations Density 98.694%