INDEX

Explanations

firm letter

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

\,.

1.23

 much

1.11

 quite

1.11

ponds

1.09

 seemed

1.08

 вроде

1.07

пси

1.04

foresaid

1.04

.}}

1.04

 pairs

1.02

POSITIVE LOGITS

 경우에는

1.47

 możli

1.35

 délégation

1.33

সজ্জিত

1.31

囷

1.30

 Geschäfts

1.29

場合は

1.28

ক্ষিপ্ত

1.27

ającego

1.26

 otomatis

1.22

Activations Density 0.000%