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

 texts

-0.43

 textes

-0.39

heten

-0.39

 Texts

-0.38

likler

-0.37

テキスト

-0.36

losen

-0.36

Su

-0.36

<eos>

-0.36

↵

-0.35

POSITIVE LOGITS

GEBURTSDATUM

0.75

rungsseite

0.72

ValueStyle

0.70

enterOuterAlt

0.70

存于互联网档案馆

0.67

 disambiguazione

0.66

SharedCtor

0.66

 CreateTagHelper

0.66

 EconPapers

0.66

 kasarigan

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.