INDEX

Explanations

progresses, reduction, over, money, owe, before, for

code and technical explanations

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ید

0.52

یا

0.50

لی

0.50

Attractive

0.49

</h4>

0.47

 Premises

0.47

Surname

0.47

College

0.46

Telephone

0.46

Leg

0.44

POSITIVE LOGITS

 slowly

0.48

 gelatin

0.47

 silica

0.47

 hauling

0.47

 stretchy

0.46

MFG

0.45

cretion

0.45

 slowing

0.45

 continúa

0.45

 stretching

0.44

Activations Density 0.001%