INDEX

Explanations

morphed, day, traction

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

 zahlreichen

-1.52

 einzigartigen

-1.52

</h5>

-1.46

 sicheren

-1.45

年来

-1.41

 Konkurrenz

-1.39

 direkten

-1.38

"));

-1.37

 unmittel

-1.36

 Hinblick

-1.35

POSITIVE LOGITS

(￣

1.32

1.30

dve

1.28

 incluse

1.25

 leves

1.25

 materi

1.25

bof

1.22

bten

1.20

 samh

1.20

 efic

1.16

Activations Density 0.553%