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

 للاسماء

-0.67

انگلیسی

-0.52

 صوتيه

-0.51

 يتيمه

-0.45

لیس

-0.43

الحياه

-0.43

 sauve

-0.40

řízení

-0.37

 league

-0.37

 seal

-0.37

POSITIVE LOGITS

itm

2.09

بالإنجليزية

1.48

期刊论文

1.25

 autorytatywna

0.95

AndEndTag

0.71

 AssemblyCulture

0.71

__(/*!

0.71

 iconLine

0.70

 estekak

0.65

MigrationBuilder

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.