INDEX

Explanations

licenses and code structure

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

as

-2.59

 Although

-2.39

if

-2.31

at

-2.25

 even

-2.23

for

-2.22

 before

-2.09

 while

-2.08

 with

-2.06

 although

-2.06

POSITIVE LOGITS

 trám

2.05

 lauded

1.90

 ſou

1.85

 astonishing

1.80

 amazingly

1.80

trying

1.77

 astounding

1.77

 acclaimed

1.77

 wondrous

1.76

 goofy

1.73

Activations Density 0.052%