INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 estekak

-0.79

 snippetHide

-0.73

TypedDataSet

-0.70

艋

-0.69

 noDo

-0.69

AsUp

-0.67

 NSCoder

-0.66

 мәкал

-0.65

 kasarigan

-0.63

ReusableCell

-0.63

POSITIVE LOGITS

 âgé

0.33

 parís

0.30

the

0.29

 approximately

0.29

 plong

0.29

 menudo

0.28

 caractères

0.28

 coute

0.27

 numerous

0.27

Hozzáférés

0.27

Activations Density 0.000%

No Known Activations

This feature has no known activations.