INDEX

Explanations

flat

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

 flat

-0.82

 FLAT

-0.81

Flat

-0.77

 Flat

-0.73

FLAT

-0.68

flat

-0.66

 flatter

-0.60

 flatten

-0.58

 flattening

-0.55

BeginContext

-0.53

POSITIVE LOGITS

 насељу

0.63

 surla

0.57

mybatisplus

0.57

клопе

0.54

spill

0.53

capac

0.51

pay

0.51

fiel

0.50

dal

0.50

anyuan

0.50

Activations Density 0.176%