INDEX

Explanations

Military and conflict

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

:variables

-0.27

 brackets

-0.26

uggage

-0.26

 Ð²ÑĭÐ¿ÑĥÑģÐº

-0.25

ergus

-0.25

ä¸įæĽ¾

-0.24

<LM

-0.24

 iParam

-0.24

éĹ®éĹ®

-0.24

meld

-0.24

POSITIVE LOGITS

å¼ĢçĽĺ

0.28

Hy

0.28

å¼ĢåŃ¦

0.26

è¶Ĭå¤§

0.26

 north

0.26

åĢĴ

0.25

 Frid

0.25

å¹´åºķåīį

0.25

 near

0.24

 loaded

0.24

Activations Density 0.003%