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

Personendaten

-0.65

 BoxFit

-0.59

UserScript

-0.56

новниш

-0.52

AddHtmlAttribute

-0.51

HideFlags

-0.50

RegressionTest

-0.48

SharedCtor

-0.48

 Normdatei

-0.47

⤹

-0.47

POSITIVE LOGITS

 iconFacebook

0.40

EndInit

0.39

 ProtoMessage

0.38

 autorytatywna

0.37

Παραπομπές

0.37

 abond

0.36

("")]

0.35

 sûre

0.35

 ***!

0.35

'{@

0.35

Activations Density 0.000%

No Known Activations

This feature has no known activations.