INDEX

Explanations

test, tests, testing, benchmark

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

is

-0.91

 是否

-0.87

 heures

-0.84

 caseros

-0.83

是否

-0.81

 rése

-0.79

девушка

-0.78

řad

-0.72

 sommes

-0.70

زینه

-0.70

POSITIVE LOGITS

 test

1.78

 tests

1.52

 testing

1.39

测试

1.34

junit

1.34

 Test

1.24

 tested

1.22

測試

1.20

 теста

1.14

 Tests

1.09

Activations Density 0.006%