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

oredCriteria

-1.03

principalColumn

-0.79

 уж

-0.47

LookAnd

-0.45

 maternidade

-0.45

canActivate

-0.44

ThemeOverlay

-0.42

 lapin

-0.41

MessageOf

-0.40

RoleName

-0.39

POSITIVE LOGITS

beginTransaction

3.55

 ContentValues

0.60

openSession

0.57

 Beginnings

0.54

 Beginning

0.54

 beginning

0.54

 وتسجيلات

0.54

migrationBuilder

0.51

 combineReducers

0.50

 chi̍t

0.49

Activations Density 0.000%

No Known Activations

This feature has no known activations.