INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

testify

-0.50

UNRELATED

-0.45

 Huguen

-0.35

finish

-0.35

VertexAttrib

-0.34

IGraphics

-0.34

Nast

-0.33

 Tarn

-0.33

 Wikipédia

-0.33

balah

-0.32

POSITIVE LOGITS

 насељу

1.46

slidesPer

1.43

дз

1.34

}{@

1.32

jfree

1.21

styleType

1.05

 يتيمه

0.81

sizeCache

0.80

 disambiguazione

0.79

ViewImports

0.73

Activations Density 0.000%

No Known Activations

This feature has no known activations.