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

FilterChain

-0.38

lag

-0.33

 sentenza

-0.33

stwie

-0.33

 său

-0.32

IPV

-0.31

 béné

-0.30

living

-0.30

 soumis

-0.30

itev

-0.30

POSITIVE LOGITS

fortawesome

5.28

 فريبيس

0.93

 AssemblyCompany

0.93

 kaarangay

0.92

IntoConstraints

0.89

ftagPool

0.89

LookAnd

0.87

NameInMap

0.85

 صوتيه

0.85

elemField

0.82

Activations Density 0.000%

No Known Activations

This feature has no known activations.