INDEX

Explanations

visual depictions of characters or teams

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

if

0.71

 applicable

0.62

for

0.62

 deleterious

0.61

If

0.58

 useful

0.57

 required

0.56

 incremental

0.55

applicable

0.55

不少

0.54

POSITIVE LOGITS

 figurines

1.29

 figurine

1.28

 silhouettes

1.27

 depicts

1.27

 portraits

1.23

 depicting

1.22

 depicted

1.22

 closeup

1.20

 silhouette

1.19

 cartoon

1.18

Activations Density 1.495%