INDEX

Explanations

conditional, shocking

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 unconditional

-1.66

 unconditionally

-1.27

conditional

-1.05

conditionally

-0.84

 conditional

-0.70

 conditionally

-0.67

Conditional

-0.64

 Conditional

-0.54

*/,

-0.52

]--;

-0.51

POSITIVE LOGITS

featureID

0.68

EndContext

0.62

ed

0.57

LAHOMA

0.54

 EClass

0.54

出版年

0.53

XmlRootElement

0.52

AnchorStyles

0.51

 Venn

0.48

shire

0.48

Activations Density 0.005%