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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kö

-0.41

cel

-0.40

 ultras

-0.36

 styl

-0.36

Collision

-0.36

sty

-0.36

 liqu

-0.36

ListFragment

-0.35

 indis

-0.35

pou

-0.35

POSITIVE LOGITS

AddTagHelper

0.65

 وتسجيلات

0.65

ValueStyle

0.64

 BorderSide

0.63

AndEndTag

0.63

elemField

0.57

 estekak

0.57

__":

0.57

__':

0.57

expandindo

0.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.