INDEX

Circuit Sparsity

No connected neurons data available

Explanations

`>` or `return` or `,` or `(` (negative activations)

path.join or path.splitext

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

reshold

-0.09

 iter

-0.08

ateg

-0.07

 steps

-0.06

 while

-0.06

_step

-0.05

 step

-0.05

run

-0.05

und

-0.05

 population

-0.05

POSITIVE LOGITS

+=

0.09

_sum

0.04

 parent

0.04

sum

0.04

ness

0.03

ge

0.03

us

0.02

0.01

ax

0.01

Activations Density 58.224%