INDEX

Explanations

phrases indicating the presence of specific objects or features associated with items

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

possessing or including

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

Alice

-0.60

 zwiſchen

-0.57

ItemList

-0.57

 yourselves

-0.57

ſelben

-0.56

niksi

-0.56

௃

-0.56

ItemModel

-0.56

 Alice

-0.55

äler

-0.55

POSITIVE LOGITS

its

0.40

 sahip

0.33

帖最后由

0.33

addCriterion

0.31

urtstag

0.29

 kehilangan

0.29

 mít

0.29

 posiada

0.27

 demikian

0.27

它的

0.26

Activations Density 0.090%

phrases indicating the presence of specific objects or features associated with items

possessing or including

No Comments

No Known Activations