INDEX

Explanations

mentions of the need for more research or analysis.

calling for future research

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

Negative Logits

 need

-2.95

need

-2.70

 Need

-2.58

Need

-2.53

 needs

-2.22

 NEED

-2.14

 Needs

-2.05

needs

-2.02

NEED

-2.00

Needs

-1.95

POSITIVE LOGITS

Nio

0.53

ExecuteReader

0.49

ωση

0.49

BindingSource

0.49

LEncoder

0.49

tamento

0.48

 OnTriggerEnter

0.48

ubri

0.47

 Bancroft

0.47

dillos

0.47

Activations Density 3.183%