INDEX

Explanations

and what

starting response phrases

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

 också

0.57

 देखील

0.55

 myös

0.54

also

0.52

 కూడా

0.50

 ankaŭ

0.50

 également

0.49

 también

0.49

 പോലും

0.49

 também

0.48

POSITIVE LOGITS

 cobre

0.43

 θέ

0.42

 هذ

0.42

铅

0.42

̂

0.41

 πε

0.40

ભાગ

0.40

'[

0.39

 amazed

0.38

Activations Density 0.002%