INDEX

Explanations

references to gender, particularly focused on men and women

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

tagHelperRunner

-0.57

 typelib

-0.52

Beat

-0.49

 onCreateView

-0.49

migrationBuilder

-0.49

AutoScale

-0.47

Kern

-0.47

 Osiris

-0.47

kern

-0.46

DockStyle

-0.46

POSITIVE LOGITS

 women

0.57

 ladies

0.54

 woman

0.53

 girls

0.48

 feminine

0.48

 female

0.46

 hembra

0.46

 females

0.44

women

0.42

 femmes

0.41

Activations Density 0.051%

references to gender, particularly focused on men and women

No Comments

No Known Activations