INDEX

Explanations

dichotomy

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

 dichotomy

-0.86

queryInterface

-0.75

assination

-0.73

 vPvB

-0.68

 cherchés

-0.65

Sucesor

-0.62

 comuniques

-0.61

 sumpay

-0.60

 שוליים

-0.59

AutoScale

-0.58

POSITIVE LOGITS

 hende

0.50

ogu

0.49

otas

0.48

 coel

0.48

ApiModelProperty

0.47

owicz

0.47

WriteAttribute

0.46

 sociali

0.45

 eventType

0.44

вге

0.43

Activations Density 0.019%