INDEX

Explanations

explanations and code improvements

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

irà

1.41

rensa

1.27

urazione

1.27

ándolo

1.25

érons

1.25

lepší

1.22

ledi

1.21

ırı

1.20

 alın

1.18

kyt

1.17

POSITIVE LOGITS

 discontent

1.59

 これらの

1.52

Согласно

1.44

vex

1.39

Despite

1.38

 이러한

1.37

 سلاٹ

1.36

Selain

1.35

 tumultuous

1.31

 haphazard

1.31

Activations Density 0.799%