INDEX

Explanations

medical journal titles

academic journal names in medical and scientific publications.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

.~(\

-1.27

 chré

-1.15

ׇ

-1.12

чком

-1.11

 teolog

-1.05

 that

-1.05

 掃除

-1.04

and

-1.02

 стекла

-1.01

lllllllll

-1.00

POSITIVE LOGITS

ּוֹ

1.44

ֶּ

1.43

 бампер

1.38

ַּ

1.37

ִּ

1.30

 patients

1.22

kowa

1.20

ֵּ

1.17

 scris

1.15

 aujourd

1.14

Activations Density 0.010%