INDEX

Explanations

predicting subsequent words

states and qualities

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 outsourcing

0.28

 τὸν

0.26

 달리

0.25

 inşa

0.25

 ارتفاع

0.25

 περισσότε

0.24

 గు

0.24

 Paryayvachi

0.24

 호텔

0.24

 церков

0.24

POSITIVE LOGITS

and

0.31

it

0.30

and

0.29

 existem

0.26

๒

0.26

及び

0.25

oui

0.25

及

0.25

ร้อม

0.25

 или

0.25

Activations Density 0.441%