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

 estekak

-0.85

zzleHttp

-0.82

 SqlConnection

-0.61

HtmlAttribute

-0.61

 Италијани

-0.55

migrationBuilder

-0.55

tagHelperRunner

-0.55

 pinulongan

-0.53

Datuak

-0.53

ScopeManager

-0.53

POSITIVE LOGITS

writeField

0.40

dodge

0.40

GIS

0.40

Draft

0.40

bab

0.39

 checksum

0.39

primary

0.37

check

0.36

 Check

0.36

ographique

0.36

Activations Density 0.000%

No Known Activations

This feature has no known activations.