INDEX

Explanations

metrics and statistics related to popularity and engagement

oai_token-act-pair · gpt-4o-mini Triggered by @bot

millions of visits or searches

np_acts-logits-general · gemini-2.5-flash-lite

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

Personendaten

-0.91



-0.73

 ब्रेकडाउन

-0.65

ंदीखरीदारी

-0.64

Insee

-0.63

BagConstraints

-0.63

\{\\

-0.59

الحياه

-0.59

 autorytatywna

-0.58

quelize

-0.56

POSITIVE LOGITS

 worldwide

0.42

 popularity

0.36

 millions

0.33

 streams

0.33

 Youtube

0.32

 internet

0.30

 massive

0.29

HasColumnName

0.29

 YouTube

0.29

 videos

0.29

Activations Density 0.002%

metrics and statistics related to popularity and engagement

millions of visits or searches

No Comments

No Known Activations