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

 OMITBAD

-0.45

 <=",

-0.45

UrlResolution

-0.44

webElementXpaths

-0.43

RTLD

-0.40

<bos>

-0.40

 Walkover

-0.39

GEBURTS

-0.38

StoryboardSegue

-0.36

istoitu

-0.36

POSITIVE LOGITS

 فريبيس

0.60

ValueStyle

0.60

twimg

0.50

 getSize

0.49

 jPanel

0.48

lblNewLabel

0.45

 CURIAM

0.45

Referencies

0.45

setPreferredSize

0.45

setViewportView

0.44

Activations Density 0.000%

No Known Activations

This feature has no known activations.