INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

usammen

-0.47

Réalisation

-0.36

NSIndexPath

-0.36

ι

-0.35

ships

-0.34

ArgumentParser

-0.33

örgy

-0.33

som

-0.32

 coal

-0.32

 conte

-0.32

POSITIVE LOGITS

 محفوظة

0.63

 typelib

0.58

WebVitals

0.56

 humanidade

0.52

OrNil

0.49

ValueStyle

0.49

 '\\;'

0.48

Datuak

0.48

PostExecute

0.47

 abordagem

0.46

Activations Density 0.000%

No Known Activations

This feature has no known activations.