INDEX

Explanations

end of instruction

completion, success, or result

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 zudem

0.54

 terlebih

0.49

 außerdem

0.47

しかも

0.46

 moreover

0.46

更是

0.45

Moreover

0.44

そも

0.44

そもそも

0.44

 speziell

0.43

POSITIVE LOGITS

 Voilà

1.23

 voila

1.14

 voilà

1.12

 Repeat

1.03

Voilà

1.02

 repeat

0.95

Repeat

0.92

 Afterwards

0.90

 afterwards

0.89

 Done

0.89

Activations Density 0.235%