INDEX

Explanations

the word "well."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_19/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.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

ління

-0.54

-0.50

これも

-0.50

GEBURTSDATUM

-0.47

 تعدى

-0.46

SourceChecksum

-0.46

pertino

-0.46

iyor

-0.45

apad

-0.45

brata

-0.45

POSITIVE LOGITS

存于互联网档案馆

0.74

دانشنامهٔ

0.68

saraba

0.66

 resourceCulture

0.66

ViewImports

0.65

Datuak

0.65

as

0.60

istoitu

0.60

ⓧ

0.58

FontOfSize

0.58

Activations Density 0.080%

the word "well."

No Comments

No Known Activations