INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.18.hook_mlp_out

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

لينكات

-0.65

 autorytatywna

-0.63

Rohy

-0.53

 MaterialApp

-0.52

adaptiveStyles

-0.52

 polaire

-0.46

 fournisse

-0.46

Meik

-0.44

 mères

-0.44

 diretta

-0.43

POSITIVE LOGITS

 незавершена

0.76

RenderAtEndOf

0.65

 unknownFields

0.56

 numerus

0.56

 OMITBAD

0.54

yntaxException

0.51

expandindo

0.49

CHAT

0.49

obu

0.49

urlencoded

0.49

Activations Density 0.000%

No Known Activations

This feature has no known activations.