INDEX

Explanations

after punctuation or verbs

historically inaccurate, physically unrealistic

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 بونس

0.48

 설립

0.48

 обре

0.47

 drows

0.47

 آہستہ

0.47

بعد

0.46

している

0.46

 തൂ

0.45

 حیر

0.45

 hereinafter

0.44

POSITIVE LOGITS

0.51

↵↵

0.44

ুর

0.42

 completos

0.41

 partes

0.40

Sc

0.40

 Tear

0.38

மக

0.38

ίζεται

0.36

完整的

0.36

Activations Density 0.000%