INDEX

Explanations

I know/you know

This neuron detects first-person plural statements, especially uses of “we” (e.g. “we know,” “we don’t,” “we can,” etc.).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

看出

0.71

 لاحظ

0.68

 देखिएगा

0.62

 보세요

0.59

を見ると

0.59

 판단

0.57

感受到

0.56

 দেখিলাম

0.55

 चैप्टर्स

0.55

 ترى

0.54

POSITIVE LOGITS

 conhe

1.73

 conhecer

1.59

 conocer

1.57

 conoce

1.57

 conocen

1.52

 conoc

1.52

 conosce

1.51

 conhece

1.50

รู้จัก

1.50

 conoz

1.48

Activations Density 0.279%