INDEX

Explanations

fleeing, escaping, or leaving

escape and avoid

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

maier

0.42

酽

0.41

 коюмдар

0.40

wość

0.40

 পুষ

0.39

ストラップ

0.39

agnet

0.38

 ഷോ

0.38

 ампли

0.38

Reusable

0.37

POSITIVE LOGITS

 fleeing

3.23

 flee

3.09

 fled

2.97

逃

2.97

 fugitive

2.58

fug

2.55

 fuga

2.53

 escaping

2.39

 escape

2.36

 পালিয়ে

2.33

Activations Density 0.100%