INDEX

Explanations

Non-English words

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 MaterialApp

-0.57

ymal

-0.56

tagHelperRunner

-0.56

InputTagHelper

-0.52

Хьажоргаш

-0.52

Ligações

-0.52

enetic

-0.51

تقاوى

-0.51

zow

-0.50

 White

-0.49

POSITIVE LOGITS

 disambiguazione

0.63

rungsseite

0.61

 שוליים

0.58

 createSlice

0.56

 carnet

0.55

grades

0.51

MCM

0.51

 IMAP

0.50

grade

0.50

SCM

0.49

Activations Density 0.004%

No Known Activations