INDEX

Circuit Sparsity

No connected neurons data available

Explanations

comparison with value

equality comparison (negative activations)

Auto-Interp Tools

Configuration

Prompts (Dashboard)

16,384 prompts, 256 tokens each

Dataset (Dashboard)

neuronpedia/python-code-simplified

Embeds

IFrame

Link

Negative Logits

_coordinates

-0.06

ry

-0.05

ri

-0.05

ordinate

-0.05

ordinates

-0.04

_coords

-0.03

 colle

-0.03

param

-0.03

umn

-0.02

(num

-0.02

Positive Logits

 part

0.10

_part

0.09

_p

0.07

art

0.06

==

0.06

mpty

0.05

(p

0.05

pi

0.05

ĥ

0.04

Activations Density 100.000%