INDEX

Explanations

Variances detected, explanation, requirement, less, restriction

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 religieuses

-2.09

 religieux

-1.92

 растений

-1.89

ׇ

-1.88

-1.83

 chré

-1.63

 מים

-1.63

 roślin

-1.57

 összefoglal

-1.55

 municípios

-1.54

POSITIVE LOGITS

ּוֹ

2.52

וּ

2.38

 бампер

2.16

 капю

2.08

᾽

1.91

 чехол

1.87

҉

1.84

ַּ

1.79

or

1.77

 your

1.77

Activations Density 0.280%