INDEX

Explanations

data generation and samples

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

፨

0.48

HALF

0.46

half

0.46

klar

0.43

둠

0.43

roweak

0.41

gn

0.40

udahkan

0.40

aliśmy

0.40

eltzer

0.40

POSITIVE LOGITS

 compliment

0.39

 efficiency

0.38

 والص

0.37

 spac

0.37

 Disbursement

0.37

 hasty

0.37

 expectation

0.36

 strained

0.36

 desire

0.35

 soff

0.35

Activations Density 0.007%