INDEX

Explanations

code and programming

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

bootstrapcdn

-0.63

 CanadaChoose

-0.63

 ContentValues

-0.61

 devriez

-0.59

protoimpl

-0.57

Παραπομπές

-0.57

+#+#

-0.57

uttavia

-0.57

 disambiguazione

-0.57

 Substances

-0.56

POSITIVE LOGITS

 good

1.13

good

0.93

 Good

0.89

Good

0.88

 GOOD

0.81

GOOD

0.77

 enough

0.63

клопе

0.55

 gute

0.53

 Geraadpleegd

0.51

Activations Density 0.015%