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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 serez

-0.29

 nocturna

-0.28

ależ

-0.28

 autorità

-0.28

 adelante

-0.27

mær

-0.27

落

-0.27

źni

-0.26

betrieb

-0.26

page

-0.26

POSITIVE LOGITS

 CommonModule

6.06

CommonModule

1.98

 BrowserModule

1.83

AnimationsModule

1.33

IntoConstraints

0.96

 TestBed

0.95

ंदीखरीदारी

0.94

providedIn

0.94

 NgModule

0.87

 ExecuteAsync

0.85

Activations Density 0.000%

No Known Activations

This feature has no known activations.