INDEX

Explanations

terms and phrases related to operational organization and structure

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

specific entities or domains

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

 nehé

-0.35

 Baumwolle

-0.31

 Navidad

-0.30

Odkazy

-0.29

baiki

-0.29

seende

-0.29

Оп

-0.29

 muñeca

-0.28

dientemente

-0.28

 Retrieved

-0.28

POSITIVE LOGITS

:✨

0.73

iſchen

0.71

uxxxx

0.68

niſſe

0.67

Personendaten

0.66

ésultats

0.65

المناصب

0.64

 zwiſchen

0.64

tableFuture

0.64

bootstrapcdn

0.64

Activations Density 0.054%

terms and phrases related to operational organization and structure

specific entities or domains

No Comments

No Known Activations