INDEX

Explanations

statistical measures and comparisons of frequency or deviation within datasets

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

difference between

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

GEBURTSDATUM

-0.62

writeFieldEnd

-0.53

 feroit

-0.52

 transfieras

-0.50

inqui

-0.49

 étoient

-0.49

 تعدى

-0.48

SourceChecksum

-0.47

 Photocase

-0.46

 disambiguazione

-0.46

POSITIVE LOGITS

TabStop

0.45

 referrerpolicy

0.41

->___

0.39

zzlies

0.33

EE

0.32

Халык

0.32

ModelBuilder

0.31

庁

0.31

ViewInit

0.31

 ویکی‌پدیای

0.31

Activations Density 0.302%

statistical measures and comparisons of frequency or deviation within datasets

difference between

No Comments

No Known Activations