INDEX

Explanations

Case in, once in

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

 calefacción

-1.05

 respiración

-0.90

mici

-0.90

 Invocation

-0.89

inches

-0.89

 personalización

-0.88

 любую

-0.86

Inches

-0.82

 PREFIX

-0.80

Dans

-0.79

POSITIVE LOGITS

in

3.86

In

1.75

In

1.23

in

1.19

în

1.17

 در

1.16

IN

0.98

 σε

0.93

and

0.88

insegna

0.87

Activations Density 0.017%