INDEX

Explanations

refer

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

lom

-0.65

mendous

-0.49

Lom

-0.49

 Chriftian

-0.48

blest

-0.48

 насељу

-0.47

 Reſ

-0.47

 ſche

-0.46

 purpoſe

-0.45

Lom

-0.45

POSITIVE LOGITS

WriteBarrier

0.57

detectChanges

0.57

قه

0.54

ⓧ

0.54

istice

0.53

TemporalType

0.53

 kiin

0.52

-};

0.51

httphttps

0.51

urti

0.51

Activations Density 0.011%