INDEX

Explanations

you"d" / "you" followed by "want", "would", "like"

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

 wanting

1.38

 want

1.25

 muốn

1.21

ต้องการ

1.21

want

1.20

Want

1.20

想要

1.19

 Want

1.18

 ต้องการ

1.12

 хотят

1.10

POSITIVE LOGITS

(++

0.33

✵

0.32

ufer

0.31

 учет

0.31

τιο

0.31

ӳ

0.31

 accumulate

0.30

 bhan

0.30

✦

0.30

astern

0.30

Activations Density 0.020%