INDEX

Explanations

thanks

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 thanks

-1.13

 thank

-0.95

 Thanks

-0.94

Thanks

-0.93

 gracias

-0.93

thanks

-0.93

 grazie

-0.85

Thank

-0.84

 Thank

-0.83

thank

-0.81

POSITIVE LOGITS

 itſelf

0.86

 raiſ

0.81

ſelves

0.79

 myſelf

0.77

 ſhe

0.77

 ་་

0.72

ſelf

0.70

 themſelves

0.69

 whoſe

0.69

 himſelf

0.68

Activations Density 0.008%