INDEX

Explanations

phrases indicating habitual actions or experiences

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iu

-0.06

LB

-0.06

indr

-0.06

xl

-0.06

 memory

-0.06

ories

-0.06

assage

-0.06

rg

-0.06

 hopes

-0.06

bla

-0.06

POSITIVE LOGITS

 being

0.09

 seeing

0.08

 hearing

0.07

Ð°ÑĢÐ°

0.07

being

0.07

ChangedEventArgs

0.07

/welcome

0.07

ãĥ³ãĤ¯

0.07

istrovstvÃŃ

0.07

poon

0.06

Activations Density 0.007%