INDEX

Explanations

elements related to photography or visual content

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

Non-English text or foreign words

np_acts-logits-general · gemini-2.0-flash

dates and places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.30.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

-0.45

OrWhiteSpace

-0.38

luß

-0.33

pageX

-0.33

providedIn

-0.32

bildēt

-0.31

NewUrlParser

-0.30

beginPath

-0.30

 ferons

-0.29

Ɓ

-0.28

POSITIVE LOGITS

 resourceCulture

0.67

 tersebut

0.63

 này

0.60

SerializedSize

0.59

 ویکی‌پدیای

0.57

টি

0.56

 autorytatywna

0.56

Parcelize

0.54

存于互联网档案馆

0.54

AnchorTagHelper

0.52

Activations Density 0.015%

elements related to photography or visual content

Non-English text or foreign words

dates and places

No Comments

No Known Activations