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

setTypeface

-0.40

ViewById

-0.35

<eos>

-0.33

FailureListener

-0.33

 Eind

-0.33

AxisAlignment

-0.33

 verv

-0.33

setContentType

-0.33

 sár

-0.32

ulism

-0.32

POSITIVE LOGITS

'{@

5.03

>{@

1.41

 autorytatywna

0.90

 nahilalakip

0.82

 AssemblyTitle

0.81

 "..\..\..\

0.74

Autoritní

0.73

EDEFAULT

0.70

 resourceCulture

0.69

ThroughAttribute

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.