INDEX

Explanations

contractions and negations

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

diary

-0.98

実に

-0.85

deutung

-0.82

之为

-0.81

 Kühlschrank

-0.81

脘

-0.81

であった

-0.79

 Youtube

-0.79

Ông

-0.79

lanta

-0.78

POSITIVE LOGITS

isn

1.38

 aren

1.13

 wasn

1.13

Couldn

1.11

 couldn

1.08

 doesn

1.08

couldn

1.04

Doesn

1.01

Didn

1.00

 didn

0.98

Activations Density 0.055%