INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

 kano

-0.48

CLS

-0.48

 advoc

-0.48

 graphique

-0.47

 Bolsonaro

-0.47

Duane

-0.47

 utili

-0.46

Kesimpulan

-0.46

Especificación

-0.45

Pfalz

-0.45

POSITIVE LOGITS

 CreateTagHelper

0.59

tagHelperRunner

0.58

 تضيفلها

0.55

 مرئيه

0.55

GOTREF

0.49

 ujednoznacz

0.48

hyrchwyd

0.48

 ModelExpression

0.47

 كومونز

0.47

OGND

0.46

Activations Density 0.000%

No Known Activations

This feature has no known activations.