INDEX

Explanations

compare

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

 Comparison

-0.65

XmlAccessorType

-0.63

 Compare

-0.61

 resourceCulture

-0.60

 comparação

-0.59

aspectj

-0.59

 semelhantes

-0.57

 comparación

-0.56

 compare

-0.54

 comparison

-0.54

POSITIVE LOGITS

mtrl

0.61

 fiduci

0.60

 myſelf

0.57

bibfield

0.57

ynthetic

0.55

orithmic

0.54

 radioactivity

0.54

(")");

0.54

 vnto

0.54

DockStyle

0.53

Activations Density 0.033%