INDEX

Explanations

You are a

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

Ỷ

1.55

acakt

1.48

 regla

1.46

ließ

1.46

 kín

1.44

 jaye

1.37

goi

1.31

 تصير

1.30

 sifatida

1.30

潦

1.29

POSITIVE LOGITS

 elsewhere

1.30

 operator

1.29

 anywhere

1.29

*;

1.24

 eleven

1.24

 experts

1.22

здат

1.20

 previously

1.18

 earlier

1.17

ayas

1.17

Activations Density 0.025%