INDEX

Explanations

hadn't or haven't activated

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

انية

2.64

чина

2.61

𝑢

2.58

 attiv

2.55

കസ

2.51

 asunto

2.50

≌

2.50

প্তাহ

2.50

 inizi

2.47

 evidente

2.43

POSITIVE LOGITS

2.68

eur

2.39

eh

2.23

다면

2.19

ek

2.16

rstrip

2.11

יים

2.07

sp

2.06

spapers

2.05

tattoo

2.04

Activations Density 0.090%