INDEX

Explanations

preposition followed by specific word

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

jaket

-1.51

escuela

-1.48

wasn

-1.30

Pratique

-1.26

 célèbre

-1.24

égal

-1.23

 borracha

-1.23



-1.20

Took

-1.19

tapete

-1.16

POSITIVE LOGITS

 which

1.63

 without

1.46

at

1.45

in

1.38

 while

1.38

 should

1.37

by

1.30

or

1.30

 amidst

1.27

 using

1.27

Activations Density 0.000%