INDEX

Explanations

descriptive adjectives that characterize various objects or entities

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

descriptive adjectives

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

 cancha

-0.47

 arşiv

-0.45

 dormitórios

-0.44

 Prozesse

-0.43

 comunicaciones

-0.42

ambién

-0.41

 skóry

-0.41

 carreras

-0.40

 Fälle

-0.40

 escaleras

-0.40

POSITIVE LOGITS

 ویکی‌پدی

0.53

+#+#

0.50

:✨

0.50

FXMLLoader

0.45

EndContext

0.43

ThroughAttribute

0.43

ArgsConstructor

0.43

 faſt

0.42

abstractmethod

0.42

ollectionView

0.41

Activations Density 0.171%

descriptive adjectives that characterize various objects or entities

descriptive adjectives

No Comments

No Known Activations