INDEX

Explanations

occur during

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

 Bakker

1.28

 Registrar

1.26

IAT

1.22

<?

1.21

 Comerc

1.17

seb

1.14

たる

1.12

 Plenum

1.10

テンツ

1.09

 recours

1.08

POSITIVE LOGITS

 бом

2.03

 assassinated

1.96

 сочи

1.95

 родился

1.92

想着

1.91

harmonic

1.87

 inflated

1.87

 inflate

1.86

 gespielt

1.85

born

1.84

Activations Density 0.012%