INDEX

Explanations

appearing legitimate

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

 Refresh

1.02

 Jamboree

0.98

 দৈর্ঘ

0.94

pana

0.93

场地

0.92

Refresh

0.92

筛

0.91

 మారింది

0.91

 Changer

0.91

最大的

0.90

POSITIVE LOGITS

 hearsay

1.37

 evidence

1.36

 literature

1.35

 precedents

1.35

 scriptures

1.28

 textual

1.28

 historical

1.28

 anecdotal

1.24

 mythology

1.24

文献

1.23

Activations Density 0.332%