INDEX

Explanations

phrases starting sentences

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

è²

-0.10

try

-0.09

ennen

-0.09

esco

-0.08

 subrange

-0.08

iken

-0.08

apper

-0.08

HOST

-0.08

 Decoder

-0.08

Com

-0.08

POSITIVE LOGITS

 Firstly

0.11

 mess

0.10

 disguised

0.10

mess

0.10

ENA

0.09

enia

0.09

 beforehand

0.09

 Mare

0.09

 First

0.09

 afore

0.08

Activations Density 0.013%