INDEX

Explanations

code snippets

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

å§¿åĬ¿

-0.26

orman

-0.24

 Facial

-0.24

å±Ģ

-0.24

æ¯Ķè¾ĥé«ĺ

-0.24

èĦ¸éĥ¨

-0.24

 facial

-0.24

SIM

-0.24

ä¸įåĲĮç¨ĭåº¦

-0.24

åı¯è¾¾

-0.23

POSITIVE LOGITS

rop

0.27

ä»£è¡¨

0.27

rop

0.26

 representatives

0.25

 representative

0.25

è¦ĨçĽĸçİĩ

0.25

è¹Ī

0.24

çĺ´

0.24

çĿ¦

0.24

è°ĥçĲĨ

0.24

Activations Density 0.003%