INDEX

Explanations

research

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 jScrollPane

-0.67

 BorderRadius

-0.66

 AssemblyVersion

-0.65

UserScript

-0.60

tagHelperRunner

-0.60

CloseOperation

-0.57

 tölt

-0.56

hips

-0.54

neté

-0.54

InputBorder

-0.52

POSITIVE LOGITS

 question

0.62

 team

0.57

question

0.50

 questions

0.49

 model

0.48

 strategy

0.46

 problem

0.46

ねん

0.45

 plan

0.45

 scenarios

0.45

Activations Density 0.001%