INDEX

Explanations

the word "extra" in various contexts

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

prite

-0.07

ronic

-0.07

otine

-0.07

 Watt

-0.07

ronics

-0.07

kad

-0.07

 Ø¹ÙħÙĪÙħÛĮ

-0.07

Ø®Ø§ÙĨÙĩ

-0.06

Cad

-0.06

Ark

-0.06

POSITIVE LOGITS

ordinary

0.17

ordin

0.13

-large

0.10

ord

0.10

ORD

0.09

 ordinary

0.09

CTION

0.08

-extra

0.08

orden

0.08

ordinal

0.08

Activations Density 0.031%