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

charCodeAt

-0.32

DoubleQuotes

-0.31

ുക

-0.29

锈钢

-0.29

 пак

-0.29

SuppressLint

-0.29

ur

-0.28

ồng

-0.28

WebServlet

-0.27

 together

-0.27

POSITIVE LOGITS

bootstrapcdn

0.77

 tartalomajánló

0.71

دانشنامهٔ

0.68

ſicht

0.67

Portail

0.66

Tembelea

0.65

ésultats

0.65

 ſans

0.62

<unused51>

0.60

<unused68>

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.