INDEX

Explanations

explaining or asking why

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

==============]

2.45

ের

2.37

roh

2.17

렇게

2.13

ymes

2.12

 RtlOverlay

2.06

roga

2.02

پرس

1.93

%@",

1.91

ര

1.90

POSITIVE LOGITS

𝘭

2.78

INTRESOURCE

2.59

ীব্র

2.50

集群

2.38

elabel

2.37

leine

2.30

ленный

2.28

兘

2.27

棿

2.27

збеки

2.26

Activations Density 0.818%