INDEX

Explanations

number

The neuron activates on phrases inviting further contact or feedback (e.g. “let me know,” “contact us,” “drop me a note,” “send us an email”).

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

Administrator

0.67

내가

0.63

私が

0.60

오늘

0.58

Agricult

0.58

Ал

0.58

Character

0.57

今日

0.57

侮

0.56

Ана

0.56

POSITIVE LOGITS

of

1.07

 ofthe

0.84

 delle

0.83

 των

0.75

 ऑफ़

0.74

 của

0.73

 são

0.71

ของการ

0.71

 della

0.70

 ETFs

0.68

Activations Density 0.009%