INDEX

Explanations

different or particular instances

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

who

-1.55

 всех

-1.48

 when

-1.42

sistence

-1.42

();

-1.42

 whom

-1.41

}$,

-1.38

who

-1.36

Where

-1.36

';

-1.36

POSITIVE LOGITS

 accordance

1.36

 there

1.27

jskiej

1.23

作った

1.21

 persino

1.20



1.18

Otras

1.18

derdag

1.17

都已经

1.16

ходя

1.15

Activations Density 0.228%

No Known Activations