INDEX

Explanations

choose between options

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oslav

2.83

ož

2.69

здат

2.69

餾

2.66

вичай

2.65

ِّف

2.58

ลำ

2.55

šť

2.55

舗

2.54

ńcz

2.54

POSITIVE LOGITS

衅

2.13

ון

2.08

ட்ரா

2.07

moment

2.02

Localization

1.98

исленность

1.93

ће

1.93

deselect

1.87

selected

1.84

ੌ

1.83

Activations Density 0.732%