INDEX

Explanations

punctuation and sentence structures

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

ìłł

-0.08

ostel

-0.08

é³

-0.08

ennen

-0.08

bitrary

-0.07

>Main

-0.07

abei

-0.07

ÐµÐ²Ð¸

-0.07

upakan

-0.07

tracer

-0.07

POSITIVE LOGITS

 thus

0.09

 Accordingly

0.08

 agreement

0.08

 accordingly

0.08

thus

0.08

Thus

0.08

 agreed

0.08

 Thus

0.07

 negotiations

0.07

 Agree

0.07

Activations Density 0.052%