INDEX

Explanations

According to

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

ä¸īåŃ£åº¦

-0.20

éª¨å¹²

-0.20

å±ħå®¶

-0.20

åĶ®

-0.20

èĦĳæµ·ä¸Ń

-0.19

çº§åĪ«

-0.19

åĲĪè§Ħ

-0.19

èĢĨ

-0.19

ç»°

-0.19

çľĭçĤ¹

-0.19

POSITIVE LOGITS

 Egypt

0.24

 Egyptians

0.24

 Egyptian

0.23

@s

0.22

Egypt

0.22

éĵ¶å·Ŀ

0.21

 hardships

0.21

éĢĥç¦»

0.20

è¡½

0.20

ä»¥èī²åĪĹ

0.19

Activations Density 0.017%