INDEX

Explanations

variable name followed by closing brace

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

èse

1.10

 plates

1.10

 chol

1.04

abhavena

1.04

 reckoning

1.03

 Fatt

1.03

 chrono

0.97

 plate

0.96

ρηση

0.96

apayati

0.95

POSITIVE LOGITS

成了

1.70

Orbit

1.64

依托

1.62

salas

1.61

성의

1.58

阿里巴巴

1.57

há

1.57

 játék

1.57

 navidad

1.56

ó

1.56

Activations Density 0.000%