INDEX

Explanations

conjunctions followed by actions

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

 downregulation

0.30

 upregulation

0.26

 elution

0.26

 invertible

0.25

 foodie

0.25

 encapsulation

0.25

 sailboat

0.25

 tunnelling

0.24

 começa

0.24

POSITIVE LOGITS

 provide

0.47

provide

0.42

 entertain

0.41

 create

0.40

 engage

0.38

 tolerate

0.36

 promote

0.36

 develop

0.36

 withstand

0.36

 participate

0.35

Activations Density 0.212%