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

 Infórmanos

-0.60

 myſelf

-0.55

AnimationsModule

-0.54

ChildScrollView

-0.54

liceerd

-0.53

UnusedPrivate

-0.53

 nakalista

-0.52

GEBURTSDATUM

-0.52

MethodManager

-0.51

 faſt

-0.51

POSITIVE LOGITS

providedIn

0.46

ViewFeatures

0.44

 AppBundle

0.31

 arbete

0.31

 semelhante

0.30

 vägen

0.30

wieś

0.30

 maș

0.30

 viață

0.30

 للاسماء

0.29

Activations Density 0.000%

No Known Activations

This feature has no known activations.