INDEX

Explanations

I thought I'd post

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

åįĩéĻį

-0.20

Ð¸ÑĢÐ¾Ð²

-0.19

åĨ·æ¼ł

-0.19

ople

-0.19

ipped

-0.18

Ð¸ÑĢÐ¾Ð²ÐºÐ°

-0.18

ä¾

-0.18

ä¹Ķä¸¹

-0.18

 beck

-0.18

Ð¸Ð»Ð°

-0.18

POSITIVE LOGITS

éĹ®éĹ®

0.26

ä»ĭç»įä¸Ģä¸ĭ

0.25

ç»Ļå¤§å®¶

0.25

èģĬèģĬ

0.24

äºĨè§£ä¸Ģä¸ĭ

0.24

è®©å¤§å®¶

0.24

æĭ¿åĩºæĿ¥

0.24

è¿Ļæĺ¯æĪĳ

0.23

rama

0.23

å¤§å®¶åĪĨäº«

0.23

Activations Density 0.101%