INDEX

Explanations

I'm not sure

programmed refusal of requests

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 ہمارے

0.45

 हमसे

0.44

 nécessite

0.41

الو

0.39

 наші

0.38

))(

0.38

 madad

0.37

ющи

0.36

 чыныгы

0.36

私たちの

0.36

POSITIVE LOGITS

 glad

1.25

 sorry

1.24

sorry

1.01

 guessing

1.00

glad

0.98

 grateful

0.96

 thankful

0.96

 Sorry

0.94

 sure

0.88

 surprised

0.88

Activations Density 0.251%