INDEX

Explanations

keywords followed by specific continuations

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

 crap

1.30

 flake

1.26

 fluffy

1.25

 lightly

1.24

 piss

1.23

 smelly

1.22

 silly

1.21

屑

1.20

 mods

1.19

 sneaky

1.15

POSITIVE LOGITS

ná

1.49

ỡng

1.43

تش

1.35

ánd

1.35

بین

1.34

rà

1.34

toc

1.33

 đạt

1.32

ták

1.31

től

1.31

Activations Density 0.000%