INDEX

Explanations

predicting the next word in a sequence

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

BtnUp

0.42

旃

0.41

 gant

0.40

 Stakes

0.38

 Stripe

0.37

误差

0.37

ナソニック

0.37

 Stripes

0.37

蔫

0.37

부

0.36

POSITIVE LOGITS

 abnormal

0.40

Alt

0.38

 component

0.38

Sol

0.38

sol

0.38

abor

0.36

MG

0.35

Se

0.35

 plas

0.35

arom

0.35

Activations Density 0.006%