INDEX

Explanations

action by an agent

action followed by by

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 biais

0.66

 ভাগের

0.42

 ላይ

0.42

न्तु

0.40

នៃ

0.40

 معه

0.40

与之

0.39

 используя

0.39

తిరే

0.38

ಿಕೊಂಡು

0.37

POSITIVE LOGITS

 oleh

3.19

by

3.11

 توسط

3.05

โดย

2.36

 द्वारा

2.28

 بواسطة

2.25

由

2.20

 bằng

2.11

by

2.09

 দ্বারা

2.06

Activations Density 1.158%