INDEX

Explanations

t

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ceptre

-0.47

 وسلم

-0.46

 JAXBElement

-0.46

 الوطنيه

-0.46

şam

-0.45

随

-0.41

+#+#

-0.41

 Erzb

-0.41

Kens

-0.41

شاركة

-0.40

POSITIVE LOGITS

addCriterion

0.73

 Waray

0.70

WithIOException

0.66

########.

0.63

principalTable

0.63

elemField

0.61

offsetof

0.60

:✨

0.59

kháu

0.59

omitempty

0.58

Activations Density 0.022%

No Known Activations