INDEX

Explanations

words associated with beginning or taking action towards something

oai_token-act-pair · gemini-2.0-flash

entering

np_token-act-pair-logits · gemini-2.0-flash

enter

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_16/width_16k/average_l0_10

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

MessageOf

-0.60

 judiciaire

-0.53

VersionUID

-0.51

 façons

-0.49

ياء

-0.47

latego

-0.47

 autonomie

-0.47

 Entering

-0.46

脚注の使い方

-0.45

uação

-0.44

POSITIVE LOGITS

 Vikipedi

0.62

 Вікіпе

0.62

OGND

0.60

 chi̍t

0.57

kmäler

0.56

مقاله

0.55

punch

0.54

ⓧ

0.54

GOTREF

0.54

 समीक्षक

0.54

Activations Density 5.597%

words associated with beginning or taking action towards something

entering

enter

No Comments

No Known Activations