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

tagHelperRunner

-0.77

IndentedString

-0.76

+#+#

-0.74

SourceChecksum

-0.73

RetentionPolicy

-0.73

utives

-0.72

DockStyle

-0.69

tonsoft

-0.66

MemoryWarning

-0.64

σθαι

-0.64

POSITIVE LOGITS

oog

0.49

kuuta

0.47

UnitOfWork

0.45

 EntityManager

0.45

����

0.44

 reporte

0.43

 reports

0.42

Tet

0.42

گانی

0.42

 reported

0.41

Activations Density 0.000%

No Known Activations

This feature has no known activations.