INDEX

Explanations

lists and conjunctions

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

we

-1.66

𖤍

-1.61

nitř

-1.61

toDouble

-1.59

 ก็

-1.58

わけで

-1.58

fahan

-1.53

 που

-1.52

 wszyst

-1.51

thats

-1.47

POSITIVE LOGITS

now

2.03

 actually

1.80

They

1.79

</b>

1.72

if

1.58

 getting

1.47

 家居

1.41

 always

1.41

 continued

1.41

entown

1.40

Activations Density 0.173%