INDEX

Explanations

keywords, tags, listing, credit

The neuron detects document section headings or metadata labels (e.g. “Abstract,” “Key words,” “Keywords,” “phrases,” 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

<i>

-1.33

you

-1.29

关卡

-1.27

 варианты

-1.27

不让

-1.26

 detenidos

-1.22

 aumentado

-1.21

💤

-1.20

–

-1.20

 $\&$

-1.18

POSITIVE LOGITS

 These

1.63

asjonen

1.58

 ۱۵

1.47

٦

1.45

 ۱۲

1.45

栳

1.45

۴

1.41



1.40

 Кроме

1.38

鴣

1.38

Activations Density 0.007%