INDEX

Explanations

if not or even

level of detail comparison

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

Ech

0.43

 θεω

0.43

^{*}=

0.42

 postérieures

0.42

 đường

0.42

𝑃

0.42

 sélectionnez

0.40

 rétréci

0.40

여행

0.40

ন্য

0.39

POSITIVE LOGITS

 downright

1.06

 outright

0.93

甚至

0.84

 addirittura

0.82

甚至是

0.78

 zelfs

0.68

 sogar

0.66

 thậm

0.63

 bahkan

0.62

 incluso

0.61

Activations Density 0.132%