INDEX

Explanations

elements related to structured data, such as JSON or configuration settings

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

Characters after HTML or image tags

np_acts-logits-general · gemini-2.0-flash

chrome web store url detail

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 дописавши

-0.88

QMetaType

-0.79

ValueStyle

-0.79

NameInMap

-0.78

EndGlobalSection

-0.77

 مشين

-0.76

writeField

-0.75

OGND

-0.73

 <=",

-0.73

IVEREF

-0.71

POSITIVE LOGITS

bi

0.48

0.44

 iconFacebook

0.43

 omnibus

0.43

0.42

IFY

0.40

 plata

0.40

Населення

0.39

cn

0.39

 bisous

0.39

Activations Density 0.046%

elements related to structured data, such as JSON or configuration settings

Characters after HTML or image tags

chrome web store url detail

No Comments

No Known Activations