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

 Administrativna

-0.70

 Савезне

-0.70

transQ

-0.58

 HasFactory

-0.56

blurRadius

-0.54

uxxxx

-0.50

fromnode

-0.50

 AssemblyProduct

-0.49

stateProvider

-0.49

ConstraintMaker

-0.49

POSITIVE LOGITS

Билгалдахарш

0.38

UseVisualStyle

0.37

Plug

0.36

 Plug

0.36

iecie

0.35

Menurutnya

0.34

 nahilalakip

0.34

 phal

0.34

 câte

0.34

AccessorTable

0.34

Activations Density 0.000%

No Known Activations

This feature has no known activations.