INDEX

Explanations

elements related to social criticism and moral judgments

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

that end in "ored," "our," or "ur"

np_acts-logits-general · gemini-2.0-flash

or + ing/ure/or

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 للاسماء

-0.63

rungsseite

-0.62

 estekak

-0.56

UnsafeEnabled

-0.55

ंदीखरीदारी

-0.55

 समीक्षक

-0.54

ConstraintMaker

-0.54

 resourceCulture

-0.54

 تضيفلها

-0.53

elemField

-0.53

POSITIVE LOGITS

ured

1.14

ors

1.12

uring

1.09

oring

1.07

urer

1.05

ures

0.99

ours

0.98

urers

0.94

urs

0.94

ored

0.93

Activations Density 0.348%

elements related to social criticism and moral judgments

that end in "ored," "our," or "ur"

or + ing/ure/or

No Comments

No Known Activations

elements related to social criticism and moral judgments

that end in "ored," "our," or "ur"

or + ing/ure/or

No Comments

No Known Activations