INDEX

Explanations

attached documents or images

This neuron activates on references to email attachments (e.g. “attached,” “attached is,” 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

</h2>

-1.35

 only

-1.34

更快

-1.23

 jette

-1.23

 utama

-1.21

 inilah

-1.21

 Negara

-1.20

を作る

-1.18

 déclare

-1.16

 ensuring

-1.15

POSITIVE LOGITS

 unſ

1.46

 ſou

1.43

毕竟

1.34



1.34

 diſt

1.34

}();

1.30

ｕ

1.27

 theirs

1.27

 ſtate

1.27

 gordura

1.24

Activations Density 0.037%