INDEX

Explanations

code comments and code

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

illors

1.33

jis

1.30

 Dris

1.25

៥

1.24

편

1.22

าง

1.21

third

1.20

Another

1.19

عود

1.18

aas

1.17

POSITIVE LOGITS

ロン

1.28

 nanot

1.27

 thấm

1.25

沐

1.25

ケー

1.24

륨

1.24

usetzen

1.23

/**

1.22

 mitochond

1.21

**/

1.21

Activations Density 0.000%