INDEX

Explanations

savings/invariance

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

 romantique

-0.62

 uſed

-0.60

 deportivos

-0.59

 themſelves

-0.58

uſ

-0.57

 aveug

-0.56

 tranſ

-0.56

 uſe

-0.55

rings

-0.54

 feroit

-0.54

POSITIVE LOGITS

InjectAttribute

0.60

Personensuche

0.60

 Standing

0.53

 незавершена

0.53

StoreMessageInfo

0.49

ItemBackground

0.49

TargetException

0.48

Externé

0.48

Standing

0.46

Portale

0.45

Activations Density 0.044%