INDEX

Explanations

which came first / parameter version

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

चाल

1.38

जा

1.35

帳に追加

1.32

ž

1.31

`);

1.23

宿泊

1.21

ža

1.21

フィス

1.20

ാപാ

1.20

havi

1.19

POSITIVE LOGITS

 hugs

1.08

dag

1.04

ED

1.02

 ràng

1.01

 shopping

1.00

 compras

0.98

 gitt

0.97

rid

0.97

nig

0.94

 ordering

0.94

Activations Density 0.018%