INDEX

Explanations

describing objects and conditions

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

NUMBER

0.55

čak

0.54

此

0.54

揊

0.54

 основной

0.54

kere

0.53

ഈ

0.53

 পত্রিকা

0.52

llo

0.52

pèce

0.52

POSITIVE LOGITS

0.51

It

0.48

 vender

0.48

 wouldn

0.48

 doesn

0.46

 vende

0.46

ек

0.45

 generales

0.45

 primal

0.45

’

0.45

Activations Density 0.000%