INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_1m/average_l0_107

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

1,048,576

Data Type

float32

Hook Name

blocks.12.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

 suma

-0.44

 poche

-0.43

fieldId

-0.42

 στις

-0.42

EnumValue

-0.41

tissa

-0.41

 viên

-0.39

łka

-0.39

 pada

-0.39

POSITIVE LOGITS

Efq

0.96

 autorytatywna

0.94

 Theſe

0.93

 Majefty

0.93

 disambiguazione

0.91

 Jefus

0.90

 Houſe

0.90

 Monfieur

0.86

ChildScrollView

0.85

 pleaſure

0.85

Activations Density 0.000%

No Known Activations

This feature has no known activations.