INDEX

Explanations

technical terms

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ä¸īéĥ¨

-0.07

LabelText

-0.06

usus

-0.06

acre

-0.06

DataURL

-0.06

iker

-0.06

çİ°æľīæĬĢæľ¯

-0.06

åĲĦåħ·

-0.06

arna

-0.06

nga

-0.06

POSITIVE LOGITS

å°¤åħ¶æĺ¯

0.10

å°¤åħ¶æĺ¯åľ¨

0.10

 especially

0.09

çī¹åĪ«æĺ¯åľ¨

0.09

å°¤ä¸ºéĩįè¦ģ

0.09

çī¹åĪ«æĺ¯

0.09

ä¸ºäºĨéģ¿åħį

0.09

ä¸ºäºĨä¿Ŀè¯ģ

0.08

 Especially

0.08

especially

0.08

Activations Density 0.558%