INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.23.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

DockStyle

-0.75

 kaarangay

-0.60

Попис

-0.59

FunctionFlags

-0.59

 Administrativna

-0.57

 HasFactory

-0.56

ConstraintMaker

-0.56

دانشنامهٔ

-0.55

MethodManager

-0.55

 مشين

-0.55

POSITIVE LOGITS

 cortesía

0.44

 Gestión

0.44

 drenaje

0.43

 puntuación

0.42

entraînement

0.42

 Formación

0.42

 imaginación

0.41

 presidenta

0.41

Empereur

0.41

 Betrachtung

0.41

Activations Density 0.000%

No Known Activations

This feature has no known activations.