INDEX

Explanations

all, followed by descriptions of quantity or variety

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

è¿ĺæľīä»Ģä¹Ī

-0.25

åħ¶å®ĥéĹ®é¢ĺ

-0.23

ä»Ģä¹Īåı«

-0.21

åĨįè¯´

-0.21

ä½łè¿ĺ

-0.21

ä¸ĢçĽ´æ²¡æľī

-0.20

æ¯³

-0.20

ä¸įå¾Ĺä¸įè¯´

-0.20

æīįåıĳçİ°

-0.20

-initialized

-0.19

POSITIVE LOGITS

ä¸įåĲĮç¨ĭåº¦

0.26

çĶ¨æĿ¥

0.25

åı¯èĥ½

0.25

åĲ«

0.25

ç¬¦åĲĪ

0.24

éĥ½æĺ¯

0.23

é«ĺåº¦

0.23

åıįæĺł

0.22

åĮħåĲ«

0.22

 likely

0.22

Activations Density 0.058%