INDEX

Explanations

references to men and women in various contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

"men" or "man"

np_acts-logits-general · gemini-2.0-flash

man or men

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

Brainz

-0.41

../../

-0.40

 Ashby

-0.38

$_['

-0.38

Obrigada

-0.38

 Budaya

-0.37

 Aurora

-0.37

 kaarangay

-0.37

\{\\

-0.36

erschutz

-0.36

POSITIVE LOGITS

man

0.71

ConstraintMaker

0.65

Men

0.64

MEN

0.63

men

0.62

Man

0.62

 homme

0.60

Man

0.60

man

0.60

Men

0.58

Activations Density 0.145%

references to men and women in various contexts

"men" or "man"

man or men

No Known Activations