INDEX

Explanations

do, perform, know, tried

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 tanaman

1.39

gabe

1.39

设施

1.37

 पौधे

1.37

 ADVISED

1.36

nales

1.36

 தாவர

1.35

weaver

1.35

oyote

1.33

uwe

1.33

POSITIVE LOGITS

in

1.50

 from

1.46

 during

1.43

to

1.38

for

1.33

 when

1.33

 throughout

1.30

 before

1.13

 inside

1.10

on

1.08

Activations Density 0.935%