INDEX

Explanations

descriptive language related to visual art and performance

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

preceding nouns, verbs, or adjectives

np_acts-logits-general · gemini-2.0-flash

become increasingly

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

DispatchToProps

-0.46

Manbalar

-0.45

 ioutil

-0.43

 galore

-0.42

 TextAppearance

-0.42

 fancy

-0.42

 Astron

-0.42

+:+

-0.42



-0.42

тека

-0.41

POSITIVE LOGITS

 ProtoMessage

0.50

ImageContext

0.41

protoimpl

0.40

帖最后由

0.37

parsedMessage

0.36

belea

0.35

xically

0.35

 arché

0.34

Rohy

0.33

ckså

0.33

Activations Density 0.258%

descriptive language related to visual art and performance

preceding nouns, verbs, or adjectives

become increasingly

No Comments

No Known Activations