INDEX

Explanations

numerical expressions and quantifiers related to measurement or quantity

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

Number followed by a unit of measurement

np_acts-logits-general · gemini-2.0-flash

quantifiers followed by nouns/descriptors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

ViewImports

-0.90

AndEndTag

-0.74

didSet

-0.72

ImageContext

-0.68

basicConfig

-0.66

migrationBuilder

-0.64

multirow

-0.64

DispatchToProps

-0.64

ValueGeneration

-0.64

 AssemblyCompany

-0.63

POSITIVE LOGITS

 braccia

0.52

Xna

0.47

жба

0.45

>[

0.45

Hes

0.44

 McNeil

0.42

-~

0.42

Th

0.41

[-

0.41

=_

0.41

Activations Density 0.030%

numerical expressions and quantifiers related to measurement or quantity

Number followed by a unit of measurement

quantifiers followed by nouns/descriptors

No Comments

No Known Activations