INDEX

Explanations

actions and behaviour

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



-1.53

and

-1.47

 both

-1.44

now

-1.33

 encantado

-1.30

׀

-1.30

 Люди

-1.27

طب

-1.27



-1.26

 sowohl

-1.25

POSITIVE LOGITS

 другие

1.55

 któ

1.52

很快就

1.43

 rzeczy

1.38

 terse

1.37

 ways

1.37

 используя

1.33

 atendimento

1.32

ва

1.30

Activations Density 0.105%