INDEX

Explanations

terms related to race or ethnicity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

NameInMap

-0.91

 الحره

-0.71

Personendaten

-0.69

 EconPapers

-0.68

 "..\..\

-0.62

gonic

-0.62

={({

-0.60

 consectetur

-0.60

PhysRev

-0.60

 propOrder

-0.58

POSITIVE LOGITS

 white

1.29

white

1.23

 White

1.15

White

1.12

 WHITE

1.10

 black

1.05

WHITE

1.00

白

1.00

black

0.99

 whiteColor

0.93

Activations Density 0.204%

terms related to race or ethnicity

No Comments

No Known Activations