INDEX

Explanations

being followed by state or identity

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



-2.13

🩻

-1.87

was

-1.84

 ยาว

-1.69

 jego

-1.69

 classico

-1.66

 decorados

-1.66

 artesanía

-1.65

鼇

-1.65

 versátil

-1.62

POSITIVE LOGITS

It

2.36

Being

2.08

While

1.98

Since

1.95

It

1.91

 Being

1.91

 What

1.90

 Ideally

1.87

while

1.80

ごちそうさまでした

1.80

Activations Density 0.010%