INDEX

Explanations

names

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

Han

-1.08

MessageTagHelper

-1.05

 nahilalakip

-0.96

Han

-0.95

MemoryWarning

-0.90

EndProject

-0.84

RenderAtEndOf

-0.82

han

-0.82

HAN

-0.81

 noDo

-0.79

POSITIVE LOGITS

ford

0.60

FORD

0.53

OVER

0.46

 Over

0.44

fin

0.44

 over

0.43

son

0.42

fords

0.41

krat

0.41

fin

0.40

Activations Density 0.043%