INDEX

Explanations

32

np_max-act · gemini-2.0-flash

The neuron fires on numeric tokens (digits or numbers, including version numbers and numeric constants).

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ecial

-0.08

aryl

-0.08

 lineman

-0.07

 Paladin

-0.07

 alist

-0.06

材

-0.06

oproject

-0.06

.shader

-0.06

итов

-0.06

 softly

-0.06

POSITIVE LOGITS

 COPYRIGHT

0.07

 creating

0.07

=.

0.06

nect

0.06

 disappears

0.06

 militia

0.06

.physics

0.06

-auth

0.06

ườn

0.06

igu

0.06

Activations Density 0.000%

32

The neuron fires on numeric tokens (digits or numbers, including version numbers and numeric constants).

No Comments

No Known Activations