INDEX

Explanations

verb and its completion

action followed by consequence

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 لە

0.43

 मद्देनजर

0.41

 তরফে

0.37

 నేపథ్యంలో

0.37

în

0.36

În

0.36

 jūs

0.34

 বছরের

0.34

在其

0.34

 zatem

0.33

POSITIVE LOGITS

it

0.40

它

0.38

eing

0.36

г

0.35

nonsense

0.35

this

0.35

imiento

0.34

и

0.34

pped

0.34

pse

0.34

Activations Density 0.130%