INDEX

Explanations

scientific / medical context

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ieteur

-0.52

 VIAF

-0.47

 transfieras

-0.47

 juridiques

-0.46

Распо

-0.46

 Meksiku

-0.46

OnLoad

-0.44

новниш

-0.44

اصيل

-0.44

BLL

-0.44

POSITIVE LOGITS

Composable

0.56

 Paglinawan

0.55

uter

0.51

tuer

0.50

 fairing

0.50

uters

0.50

skap

0.49

utor

0.49

æder

0.49

graphies

0.49

Activations Density 0.001%

No Known Activations