INDEX

Explanations

in, at, over, since, with, by

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

 because

-1.91

 when

-1.63

 after

-1.45

 encantó

-1.43

 trouverez

-1.26

 despite

-1.23

 karena

-1.23

<bos>

-1.23

 which

-1.20

 those

-1.19

POSITIVE LOGITS

為に

1.38

の方に

1.34

際には

1.28

to

1.27

様な

1.25

ときの

1.22

ほら

1.19

場合には

1.16

んですよ

1.16

Hmm

1.16

Activations Density 0.123%