INDEX

Explanations

contamination or unwanted materials

np_max-act · gemini-2.0-flash

This neuron activates on mentions of specific technical components or module parts (especially names given in parentheses, like “CMOS (image sensor)”).

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 gust

-0.07

 Audio

-0.06

upil

-0.06

 remar

-0.06

Oh

-0.06

Sof

-0.06

 miracle

-0.06

 Icon

-0.06

 الي

-0.06

POSITIVE LOGITS

ayscale

0.07

geh

0.07

 deploying

0.06

lessly

0.06

ensburg

0.06

 вели

0.06

);
↵
↵
↵

0.06

 fused

0.06

 policies

0.06

 пользователя

0.06

Activations Density 0.054%

contamination or unwanted materials

This neuron activates on mentions of specific technical components or module parts (especially names given in parentheses, like “CMOS (image sensor)”).

No Comments

No Known Activations