INDEX

Explanations

the act of monitoring

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

’

-2.61

 Unterschiede

-2.58

缯

-2.39

 beliebtesten

-2.33

-2.27

end

-2.20

 Behaup

-2.13



-2.13

 дают

-2.09

 hervorragend

-2.09

POSITIVE LOGITS

柃

2.56

出來的

2.39

趼

2.33

 terjadinya

2.22

);

2.14

聍

2.14

Ꮉ

2.13

鱘

2.13

辔

2.11

м

2.05

Activations Density 0.017%