INDEX

Explanations

Technical papers

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 للمعارف

-0.57

enschappe

-0.47

<0x84>

-0.45

setContentView

-0.45

Törté

-0.43

 onor

-0.43

 preghi

-0.43

Oph

-0.42

VolleyError

-0.42

 ficando

-0.41

POSITIVE LOGITS

 damp

0.59

queryInterface

0.56

 vitality

0.56

 hearth

0.56

 dampen

0.55

 MainAxisSize

0.55

CppMethod

0.54

 drinking

0.54

wet

0.54

0.53

Activations Density 0.000%