INDEX

Explanations

scientific article links

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

输

1.40

 zelfs

1.28

ग्रस्त

1.26

 учиты

1.25

輸

1.24

essen

1.20

 Lask

1.18

্যন্ত

1.17

AG

1.16

 destru

1.15

POSITIVE LOGITS

 👋

1.95

大家好

1.74

 сьогодні

1.60

iamo

1.60

 พี่

1.58

 hello

1.54

您

1.53

 greetings

1.53

Buongiorno

1.52

 今天

1.52

Activations Density 0.252%