INDEX

Explanations

Seinfeld unit testing session

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

 with

-2.25

how

-1.98

 that

-1.96

 should

-1.65

 both

-1.55

cowboy

-1.45

 fragte

-1.45

at

-1.42

 this

-1.38

 atât

-1.38

POSITIVE LOGITS

 difer

1.59

 монта

1.47

 simila

1.44

1.41

Writing

1.41

 Dessa

1.41

۷

1.40

 exag

1.38

 intrigu

1.38

 Pense

1.38

Activations Density 0.166%