INDEX

Explanations

but followed by a contrast

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

甚至

-1.34

 даже

-1.31

omo

-1.27

 حالی

-1.23

тся

-1.22

幸好

-1.19

abatic

-1.18

}}\

-1.18

um

-1.16

ској

-1.16

POSITIVE LOGITS

on

1.69

 this

1.40

 today

1.36

最初は

1.29

 vandaag

1.28

paku

1.23

ınızı

1.22

 dagelijks

1.20

 there

1.18

 Başkanı

1.17

Activations Density 0.026%