INDEX

Explanations

Software and programming

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

è°ĥåĬ¨

-0.29

lur

-0.27

 capacities

-0.25

èıĢ

-0.25

è¾Ħ

-0.25

gar

-0.25

 nouveau

-0.24

 probabil

-0.24

ç¶ļ

-0.24

 balance

-0.24

POSITIVE LOGITS

aminer

0.33

æļĤç¼ĵ

0.26

DetailsService

0.26

çĶ³åįļ

0.25

æĥ³åĥı

0.25

 ThÆ°

0.25

 shampoo

0.24

æŁ¥çľĭåħ¨æĸĩ

0.23

ummer

0.23

çº§åĪ«çļĦ

0.23

Activations Density 0.007%