INDEX

Explanations

mentions of positive and negative feedback or reviews

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

asti

-0.08

asto

-0.08

~-~-

-0.08

redirectTo

-0.07

avra

-0.07

illis

-0.07

JKLM

-0.07

Ã¼f

-0.07

æĸ½

-0.07

astos

-0.07

POSITIVE LOGITS

 attention

0.07

 response

0.06

 press

0.06

 spot

0.06

dap

0.06

iji

0.06

 review

0.06

uro

0.06

 reactions

0.05

attention

0.05

Activations Density 0.017%