INDEX

Explanations

code documentation practices

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

 sonucu

1.29

xd

1.26

udah

1.25

gae

1.25

YAN

1.23

yh

1.22

idelberg

1.20

FileWriter

1.19

 Axel

1.18

bav

1.18

POSITIVE LOGITS

 диапазо

1.54

觑

1.42

збе

1.40

的关键

1.36

Campo

1.35

 пропор

1.34

 dices

1.34

 управлять

1.32

 какая

1.30

 quell

1.29

Activations Density 0.008%