INDEX

Explanations

one / someone

This neuron activates on evaluative or opinionated statements—phrases that express praise, judgments, or recommendations.

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

এবং

0.94

 selves

0.84

 మరియు

0.76

ங்களையும்

0.75

 உள்ளன

0.75

 democracies

0.75

蟶

0.73

 padrões

0.73

 niveles

0.71

に加え

0.71

POSITIVE LOGITS

 atau

1.40

 или

1.37

 หรือ

1.34

 یا

1.30

 কিংবা

1.28

 或者

1.27

或者是

1.26

或

1.24

 किंवा

1.22

 või

1.21

Activations Density 1.044%