INDEX

Explanations

references to "girls" or "Girls"

oai_token-act-pair · gpt-4o-mini Triggered by @bot

girls and boys

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RectangleBorder

-0.57

inerary

-0.56

__*/

-0.56

 McLeod

-0.53

BibitemShut

-0.51

inerario

-0.50

HasBeenSet

-0.50

Weg

-0.50

niſſe

-0.49

 McIntosh

-0.49

POSITIVE LOGITS

 Girls

2.00

 girls

1.98

Girls

1.96

girls

1.86

 GIRLS

1.80

GIRLS

1.48

 meisjes

1.31

 boys

1.23

Boys

1.22

 Boys

1.18

Activations Density 0.005%

references to "girls" or "Girls"

girls and boys

No Comments

No Known Activations