INDEX

Explanations

numerical dimensions and measurements

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

numbers and measurements

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

 protoimpl

-0.52

 Administrativna

-0.47

lihood

-0.44

ActivityResult

-0.44

 billions

-0.44

MockBean

-0.44

 interstate

-0.43

########.

-0.43

 gede

-0.42

aroo

-0.42

POSITIVE LOGITS

 Longueur

0.42

 room

0.42

 numberOfRows

0.41

 rooms

0.40

 kitchen

0.39

 largeur

0.39

 onCreateView

0.38

 floor

0.37

complexContent

0.37

 Kitchen

0.37

Activations Density 0.000%

numerical dimensions and measurements

numbers and measurements

No Comments

No Known Activations