INDEX

Explanations

mostly related to amounts or determination

quantifiers and requests

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 Sementara

0.21

 adlı

0.20

 Merkezi

0.20

 Üniversitesi

0.19

 peringkat

0.19

 Umum

0.19

 kelahiran

0.19

 Söz

0.19

 Merkez

0.19

 Trung

0.19

POSITIVE LOGITS

書いて

0.23

 கொஞ்சம்

0.23

 최대한

0.22

少し

0.21

좀

0.21

 എല്ലാം

0.20

出来る

0.20

 خیلی

0.20

 একটু

0.20

立马

0.19

Activations Density 0.003%