INDEX

Explanations

new followed by nouns

new developments and additions

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 nových

1.15

 nieuwe

1.13

 nuevas

1.09

új

1.08

new

1.05

 novos

1.04

 nowych

0.99

 nuevos

0.98

 fresh

0.98

 പുതിയ

0.97

POSITIVE LOGITS

다

0.78

اً

0.69

сі

0.62

 algebraically

0.61

८

0.59

ている

0.57

ろ

0.57

くて

0.56

はある

0.56

بی

0.56

Activations Density 0.113%