INDEX

Explanations

positive and negative outcomes

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

iary

1.98

 مقدس

1.94

ոլ

1.92

ienie

1.91

adora

1.91

ory

1.88

жина

1.86

ownika

1.85

 الأعمال

1.85

лта

1.84

POSITIVE LOGITS

גע

1.33

GRAY

1.26

setVertical

1.23

 certains

1.19

 súa

1.19

 luat

1.18

得上

1.15

 effets

1.14

vp

1.13

 tirer

1.13

Activations Density 0.006%