INDEX

Explanations

predictions and community contexts

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

 catastrophic

1.31

aile

1.27

reminder

1.26

controle

1.26

 második

1.26

regulating

1.23

ˀ

1.23

țial

1.22

 Sommers

1.22

ministerium

1.22

POSITIVE LOGITS

 fulfils

1.43

 behaves

1.30

 อาจ

1.13

 นั้น

1.12

 undergoes

1.11

 fulfills

1.08

 might

1.07

 deserves

1.05

 could

1.05

บาท

1.03

Activations Density 0.000%