INDEX

Explanations

explicit expressions for

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 использова

-1.35

 compléter

-1.30



-1.28

を使用

-1.23

を使用して

-1.22

を選び

-1.22

嵓

-1.21

튿

-1.21

 délais

-1.21

那天

-1.21

POSITIVE LOGITS

for

2.09

 from

1.49

 with

1.44

 without

1.42

in

1.38

can

1.30

 when

1.27

 does

1.20

how

1.19

 what

1.18

Activations Density 0.084%