INDEX

Explanations

lungs

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 never

-1.02

never

-0.90

 nigdy

-0.85

Never

-0.84

 Never

-0.82

 aldrig

-0.82

 nooit

-0.81

 aldri

-0.81

 nunca

-0.80

AndEndTag

-0.79

POSITIVE LOGITS

}}/>

0.54

ovács

0.54

vosti

0.53

 Tsche

0.52

neous

0.51

 offering

0.49

 apparent

0.49

 excreted

0.48

 addressed

0.48

contentLoaded

0.47

Activations Density 0.378%