INDEX

Explanations

here or initial

speak about approximations

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

ʔ

0.46

nsan

0.45

вица

0.45

艹

0.44

💘

0.44

গ্রেস

0.43

rio

0.43

 Gölü

0.43

стка

0.42

 molécules

0.42

POSITIVE LOGITS

 colloquial

0.51

 यहाँ

0.49

 ஆரம்ப

0.48

 peculiarity

0.46

here

0.46

 propon

0.46

 here

0.45

 peculiarities

0.45

initial

0.44

 நீங்கள்

0.44

Activations Density 0.001%