INDEX

Explanations

the beginning of paragraphs or sections in a document

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

legal, feature, action, spectrum

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

 ivelany

-0.65

kozások

-0.62

 XCTest

-0.59

BeginContext

-0.57

enumi

-0.57

 článku

-0.55

expandindo

-0.54

adaptiveStyles

-0.54

WebServlet

-0.54

protoimpl

-0.53

POSITIVE LOGITS

dite

0.52

nextToken

0.45

hist

0.43

ount

0.43

kosh

0.42

 portátiles

0.41

postValue

0.41

 elect

0.39

Princip

0.39

equi

0.39

Activations Density 0.075%

the beginning of paragraphs or sections in a document

legal, feature, action, spectrum

No Comments

No Known Activations