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

 &___

-0.68

 EconPapers

-0.63

TRAILING

-0.62

 Taktlose

-0.61

 propOrder

-0.53

omitempty

-0.52

addCriterion

-0.52

 تضيفلها

-0.50

 utafitiHapana

-0.50

Personensuche

-0.49

POSITIVE LOGITS

kadot

0.37

mphony

0.37

şiv

0.36

 speranza

0.35

 hatta

0.34

以上的

0.33

ICEF

0.33

ursday

0.33

izr

0.33

 viagem

0.33

Activations Density 0.000%

No Known Activations

This feature has no known activations.