INDEX

Explanations

setting up goals

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

🐏

-0.81

ifikationer

-0.81

lede

-0.79

दा

-0.75

 Mauro

-0.73

嚴重

-0.73

 тяже

-0.73

saa

-0.73

çade

-0.72

gé

-0.71

POSITIVE LOGITS

 achievable

2.64

 realistic

2.64

 attainable

2.44

 ambitious

2.20

realistic

2.19

 lofty

2.14

Realistic

2.09

 unrealistic

2.08

 Realistic

1.95

 reachable

1.78

Activations Density 0.050%