INDEX

Explanations

question

The neuron detects mentions of “question” (and its plural/forms) in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

tablissements

-1.09

if

-0.97

 Diplo

-0.96

 kaos

-0.93

优惠

-0.90

 tono

-0.90

of

-0.89

Praxis

-0.89

 daim

-0.89

ایان

-0.88

POSITIVE LOGITS

 question

1.35

Ẩ

1.25

 épreuves

1.23

 השאלה

1.21

otechnical

1.21

 matières

1.21

bước

1.17

Ayrıca

1.16

opheles

1.16

 bénéfices

1.16

Activations Density 0.019%