INDEX

Explanations

evaluate

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Ensuring

-3.41

You

-3.22

Our

-3.11

 Could

-3.08

 Former

-2.89

 selben

-2.84

-2.73

 Allowing

-2.73

 Anyone

-2.70

 Featuring

-2.70

POSITIVE LOGITS

呖

3.02

 parachoque

2.70

 capitulo

2.70

⽌

2.70

 مختلفة

2.67

 puerto

2.63

 rubio

2.59

‪.‬‬

2.58

 花卉

2.58

 humo

2.58

Activations Density 0.032%