INDEX

Explanations

backward, logarithm, and with it

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

ر

3.81

er

3.33

erà

2.87

ه

2.84

н

2.81

チュラル

2.79

erus

2.78

omaly

2.74

رر

2.71

رخ

2.66

POSITIVE LOGITS

 flatter

2.35

ㄲ

2.32

quefois

2.27

vskip

2.25

raison

2.24

ிகை

2.19

漑

2.18

ర్వాత

2.15

<0xBE>

2.14

shake

2.13

Activations Density 0.011%