INDEX

Explanations

reducing or eliminating negatives

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

 আপাতত

0.87

 irgendwie

0.87

 agak

0.86

 tentativa

0.84

 provavelmente

0.84

 هيبقى

0.83

 semblent

0.81

 talvez

0.81

 probably

0.81

 seemed

0.80

POSITIVE LOGITS

 eliminate

1.40

 eliminates

1.37

 elimination

1.35

Elim

1.34

 eliminating

1.33

 Elimin

1.30

 Eliminate

1.29

 eliminated

1.28

elimin

1.27

 Elimination

1.26

Activations Density 0.173%