INDEX

Explanations

state set or configured

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

 oversaw

-0.99

Tested

-0.99

 Tested

-0.94

Accessed

-0.93

 Found

-0.91

 begged

-0.91

Guided

-0.86

 Provided

-0.85

 encontrada

-0.85

 Accessed

-0.84

POSITIVE LOGITS

 installed

1.98

 enrolled

1.47

 parked

1.28

 deployed

1.26

 planted

1.20

 stored

1.20

 implanted

1.19

installed

1.17

 registered

1.15

 configured

1.11

Activations Density 0.241%