INDEX

Explanations

first time doing something

The neuron detects mentions that it’s the user’s “first” time (e.g. “first time,” “first visit,” etc.) in a post.

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

 sogen

-1.41

 That

-1.34

 sogenannten

-1.29

'].

-1.27

 sogenannte

-1.25

 erfolgre

-1.23

なので

-1.22

 that

-1.21

şa

-1.20

 zodat

-1.19

POSITIVE LOGITS

臯

1.43

礎

1.25

 hierdie

1.24

 alug

1.23

 ながら

1.23

 nouveaux

1.16

 pessoais

1.13

 ज़

1.13

 делают

1.10

τουργ

1.09

Activations Density 0.043%