INDEX

Explanations

units of measurement and their uses

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

Sizes

1.27

 двух

1.27

 कई

1.23

 velič

1.20

 очень

1.20

 フル

1.19

ちょっと

1.18

 два

1.18

 وي

1.16

 وتم

1.15

POSITIVE LOGITS

apport

1.08

 bonus

1.00

 supplemental

0.95

raining

0.93

 transfer

0.92

 recovery

0.91

recovery

0.90

 residual

0.90

cend

0.90

 expertise

0.90

Activations Density 0.317%