INDEX

Explanations

instances of the word "der" in various forms and contexts, indicating a focus on a specific linguistic feature

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

van der and ter prefixes

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 bénévol

-0.51

 Inscrivez

-0.47

 Komunikasi

-0.45

 Penga

-0.44

 deuil

-0.44

 bocetos

-0.43

 Bewußt

-0.43

 então

-0.43

 coraggio

-0.42

 Ciencia

-0.42

POSITIVE LOGITS

/*

0.50

 Parcelable

0.47

GEBURTSDATUM

0.47

/**

0.45

 continúas

0.45

ung

0.44

knecht

0.44

pri

0.44

 Ader

0.43

Ril

0.43

Activations Density 0.001%

instances of the word "der" in various forms and contexts, indicating a focus on a specific linguistic feature

van der and ter prefixes

No Comments

No Known Activations