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

ख

-0.57

 kasarigan

-0.56

EndGlobalSection

-0.55

énd

-0.54

raught

-0.54

 correctes

-0.51

 Familienname

-0.50

bazel

-0.49

setTypeface

-0.49

bcryptjs

-0.49

POSITIVE LOGITS

Geplaatst

0.69

 NSCoder

0.67

<?

0.61

ArrowToggle

0.60

Hauptartikel

0.59

tonsoft

0.54

 Sarm

0.52

 demurrer

0.52

 Schengen

0.52

 Baillargeon

0.52

Activations Density 0.000%

No Known Activations

This feature has no known activations.