INDEX

Explanations

quantization, feedback, dialogue, alliances, aesthetic

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

and

1.43

 throughout

1.27

 wielding

1.23

 extravaganza

1.18

 with

1.16

 during

1.14

 offerings

1.13

for

1.13

 regardless

1.13

 beside

1.13

POSITIVE LOGITS

 obtem

1.51

 jeho

1.51

 njegov

1.51

 utilizzare

1.50

 możli

1.50

 nagyon

1.48

kében

1.48

 zwró

1.47

 maneira

1.46

 njegova

1.44

Activations Density 0.294%