INDEX

Explanations

Here's a paraphrase or translation

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

 особа

1.49

 председа

1.45

သား

1.44

ὸς

1.41

جهه

1.41

 photocurrent

1.40

{};

1.39

اول

1.37

لون

1.37

rank

1.37

POSITIVE LOGITS

pack

1.31

 phức

1.25

stit

1.21

utable

1.17

oltre

1.10

)<

1.08

 flexibles

1.06

 DISCLAIMED

1.06

除

1.05

拗

1.04

Activations Density 0.028%