INDEX

Explanations

Greetings at the start of user input

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

off

1.86

1.51

Off

1.41

 core

1.35

Met

1.28

1.27

 continued

1.26

 Sang

1.26

↵

1.23

POSITIVE LOGITS

osterol

2.16

ুগত্য

2.10

<unused1003>

2.05

<unused633>

2.04

<unused139>

2.03

ದಲ

2.00

ikke

2.00

<unused995>

1.99

<unused1224>

1.99

<unused1088>

1.99

Activations Density 0.003%