INDEX

Explanations

bulbs, lamps, and lighting fixtures

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

they

-1.56

if

-1.30

 вони

-1.26

躋

-1.24

AFFIRMED

-1.24

is

-1.23

 offrant

-1.22

ኋ

-1.22

ゼ

-1.20

らなかった

-1.19

POSITIVE LOGITS

 also

1.51

 invented

1.48

もある

1.38

on

1.35

 respected

1.31

↵↵

1.30

 fragte

1.30

vinilo

1.29

 الثالث

1.28

 wusste

1.28

Activations Density 0.028%