INDEX

Explanations

references to male characters or figures

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nakalista

-0.99

^(@)

-0.96

expandindo

-0.96

 EconPapers

-0.93

 محفوظة

-0.93

 snippetHide

-0.92

']")

-0.87

})));

-0.87

 Escolar

-0.85

 Ramadhan

-0.85

POSITIVE LOGITS

men

0.87

Men

0.79

who

0.78

Man

0.68

MEN

0.67

man

0.67

Men

0.62

men

0.62

 Woman

0.54

Man

0.53

Activations Density 0.087%

references to male characters or figures

No Comments

No Known Activations