INDEX

Explanations

No Explanations Found

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 BoxFit

-0.66

出版年

-0.53

DIPSETTING

-0.51

 كومونز

-0.50

 الاطلاع

-0.50

addPreferredGap

-0.50

GEBURTSDATUM

-0.50

pencer

-0.48

Географиясе

-0.48

AddHtmlAttribute

-0.47

POSITIVE LOGITS

+:+

0.59

umque

0.53

 Gelegenheit

0.41

 feroit

0.41

 pouvoit

0.41

+#+

0.41

 AssemblyCulture

0.41

 langt

0.40

FTFY

0.39

 Serikat

0.39

Activations Density 0.000%

No Known Activations

This feature has no known activations.