INDEX

Explanations

ordinal numbers referring to rank or sequence

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

èĢı

-0.25

ä¸ĵåĮº

-0.24

kÄ±l

-0.23

ï¸

-0.23

aygÄ±

-0.23

mÄ±z

-0.23

 ÐĶÐ¼

-0.23

SSIP

-0.23

Ð°ÑĤÑĥÑĢ

-0.22

æ±Ī

-0.22

POSITIVE LOGITS

-only

0.27

æ¬Ħ

0.26

-se

0.24

-tier

0.24

worth

0.23

-order

0.23

åĨ¶

0.23

çĽİ

0.23

orns

0.23

-row

0.23

Activations Density 0.020%