INDEX

Explanations

common nouns and articles

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

images and the photographic process

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 <>",

-0.47

 AssemblyCompany

-0.45

daille

-0.45



-0.44

出版年

-0.44

Autoritní

-0.44

 httpClient

-0.41

 כך

-0.40

addContainerGap

-0.40

Bibliograf

-0.40

POSITIVE LOGITS

rawDesc

0.56

 renfer

0.43

encodeWith

0.43

 gynhyrchwyd

0.42

 Edelstahl

0.42

 snippetHide

0.41

ViewFeatures

0.39

ButterKnife

0.39

 contiennent

0.38

CppCodeGen

0.37

Activations Density 0.007%

common nouns and articles

images and the photographic process

No Comments

No Known Activations