INDEX

Explanations

profession followed by name

The neuron detects words naming a person’s role or occupation (e.g. “author,” “producer,” “reviewer,” “musician,” “scientist,” “singer,” etc.).

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

-1.57

了一些

-1.56

-1.43

-1.42

 quando

-1.41

-1.36

了许多

-1.27

-1.26

或者是

-1.24

POSITIVE LOGITS

 extraordinaire

1.70

 лъ

1.44

 triko

1.41

🎊

1.36

無し

1.34

 semoga

1.30

 دلم

1.29

heden

1.29

 bernama

1.27

 créées

1.27

Activations Density 0.036%