INDEX

Explanations

have

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Према

-0.64

")[

-0.57

NOPQRST

-0.57

lujah

-0.56

 possam

-0.54

rowable

-0.54

 Normdatei

-0.53

allage

-0.53

lihood

-0.53

InputBorder

-0.53

POSITIVE LOGITS

 ever

0.75

 much

0.68

any

0.67

 anyone

0.61

 anything

0.57

much

0.53

had

0.53

 anybody

0.52

any

0.52

 made

0.51

Activations Density 0.334%