INDEX

Explanations

expressions related to recommendations and feedback

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

lied

-0.08

.ErrorCode

-0.08

ubby

-0.07

evin

-0.07

WithPath

-0.07

Ð¾Ð¶ÐµÑĤ

-0.07

duk

-0.07

ÏĨÎŃ

-0.07

à¸²à¸Ķ

-0.07

à¤¢

-0.07

POSITIVE LOGITS

 future

0.08

 likelihood

0.07

.future

0.06

ÏĢÎ·

0.06

Lik

0.06

 rate

0.06

 would

0.06

 purchases

0.06

 again

0.05

 purchase

0.05

Activations Density 0.002%