INDEX

Explanations

don't hesitate to ask

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

va

1.55

rl

1.47

lectoral

1.46

အပ်

1.46

вної

1.42

 호출

1.40

jším

1.39

<unused500>

1.37

ve

1.35

rman

1.34

POSITIVE LOGITS

其中

1.42

各種

1.27

pointing

1.22

 milhares

1.21

દ્ધ

1.20

 amid

1.20

 partout

1.18

▲

1.17

 tant

1.15

around

1.14

Activations Density 0.089%