INDEX

Explanations

linking words to following nouns

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 remé

-1.39

 fré

-1.35

 пишет

-1.34

 yelek

-1.30

 péro

-1.29

わけではない

-1.28

 agir

-1.26



-1.24

 emprunt

-1.23

渽

-1.23

POSITIVE LOGITS

 when

1.59

 what

1.34

as

1.30

 where

1.28

 which

1.21

 would

1.20

 only

1.20

 beide

1.19

who

1.18

 after

1.17

Activations Density 0.066%