INDEX

Explanations

less efficient

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

 devout

0.44

Everyone

0.41

 револю

0.38

 revolutionize

0.38

 reportedly

0.37

MCU

0.37

Munic

0.37

Philosophy

0.36

utable

0.35

 eminent

0.35

POSITIVE LOGITS

 identities

0.42

suffixes

0.42

 tiled

0.42

 reductions

0.41

 على

0.41

 suffixes

0.40

 tiles

0.40

 brillo

0.40

 subtraction

0.39

 estuaries

0.39

Activations Density 0.003%