INDEX

Explanations

terms related to central locations and geographic positioning

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

central locations

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

Citiți

-0.40

 étranger

-0.39

 brasileño

-0.39

 peruano

-0.37

 portata

-0.35

 arrivato

-0.35

AfterViewInit

-0.35

áss

-0.35

irinha

-0.34

Abstract

-0.34

POSITIVE LOGITS

IntoConstraints

0.79

 سكانية

0.68

sizeCache

0.63

parsedMessage

0.57

ftagPool

0.56

SharedCtor

0.55

 downtown

0.54

IsMutable

0.54

 Italijanski

0.53

DrawerToggle

0.52

Activations Density 0.008%

terms related to central locations and geographic positioning

central locations

No Comments

No Known Activations