INDEX

Explanations

references to women and their experiences or empowerment

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

"women" or "woman"

np_acts-logits-general · gemini-2.0-flash

women

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

-0.44

des

-0.44

 Sanford

-0.43

 Sykes

-0.42

roy

-0.41

 Peralta

-0.41

 Coronation

-0.41

 Stark

-0.40

Smy

-0.40

DES

-0.40

POSITIVE LOGITS

Woman

1.13

 Woman

1.09

 woman

1.09

 WOMAN

1.05

Women

1.05

 women

1.02

 Women

1.00

WOMAN

0.98

WOMEN

0.96

women

0.95

Activations Density 0.049%

references to women and their experiences or empowerment

"women" or "woman"

women

No Comments

No Known Activations