INDEX

Explanations

phrases indicating something was missed or not fully investigated, and general question-asking phrases

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 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

omalies

-0.60

 BoxDecoration

-0.59

himovic

-0.58

}))

-0.58

 snippetHide

-0.57

principalColumn

-0.56

RectangleBorder

-0.56

اعمال

-0.55

ISupport

-0.54

sizeCache

-0.54

POSITIVE LOGITS

UnknownFields

0.54

 téléphonique

0.51

 Paglinawan

0.49

 censiti

0.48

MergeFrom

0.47

 GetEnumerator

0.46

Hozzáférés

0.45

 ließen

0.44

asaan

0.44

 Driven

0.44

Activations Density 0.340%