INDEX

Explanations

pollution, pollutants, pollution-induced

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

panse

-0.09

hev

-0.08

ÙĦØ³Ø©

-0.08

undle

-0.08

åħ¶äºĭ

-0.07

weise

-0.07

ÙħØ§Øº

-0.07

äºļçī¹

-0.07

epend

-0.07

ynt

-0.07

POSITIVE LOGITS

ç¨ĭåº¦

0.09

æºĲçļĦ

0.09

rious

0.08

liness

0.07

çī©çļĦ

0.07

/error

0.07

ç¨ĭåº¦çļĦ

0.07

ä¸įåłª

0.07

ĩ

0.07

ious

0.07

Activations Density 0.029%