INDEX

Explanations

desarrollo, identificación, métodos

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

 Pember

1.30

 repeat

1.27

成绩

1.25

yte

1.25

 lanc

1.21

lt

1.20

 aniversario

1.20

fdata

1.20

 finish

1.20

iters

1.19

POSITIVE LOGITS

rakt

1.15

'_{

1.14

 مباشرة

1.11

roskop

1.08

 zeml

1.06

ᐊ

1.05

섦

1.05

滑动

1.04

ahaha

1.04

的部分

1.03

Activations Density 0.038%