INDEX

Explanations

punctuation marks and their usage, indicating discourse and conversational flow

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

bubble

-0.08

clave

-0.08

eldo

-0.07

asca

-0.07

tml

-0.07

grounds

-0.07

pNet

-0.07

tach

-0.07

egend

-0.07

iÄħ

-0.07

POSITIVE LOGITS

 first

0.06

Bod

0.06

ÑĥÐºÐ¸

0.06

 simply

0.06

izi

0.06

 Strauss

0.05

tac

0.05

 Franti

0.05

 prior

0.05

Activations Density 0.031%