INDEX

Explanations

praising/promotion

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 boasted

-0.77

 Roskov

-0.74

 boast

-0.71

UVA

-0.71

 touted

-0.70

 brag

-0.65

CppCodeGen

-0.63

"}")

-0.63

MigrationBuilder

-0.63

sted

-0.60

POSITIVE LOGITS

of

0.61

ruzzo

0.48

 của

0.48

ensement

0.48

tú

0.47

unjukan

0.46

 slutt

0.44

UCTION

0.43

 sviluppo

0.43

 biaya

0.43

Activations Density 0.167%