INDEX

Explanations

phrases related to changes and processes involving system states or commands

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

avis

-0.07

æ·»

-0.07

mlink

-0.06

kud

-0.06

lick

-0.06

 faint

-0.06

 NotImplemented

-0.06

vida

-0.06

Ø®Ø´

-0.06

 Harden

-0.06

POSITIVE LOGITS

 pent

0.15

 waiting

0.11

pent

0.11

 stored

0.10

 held

0.10

 Pent

0.10

 bottled

0.10

 Held

0.09

 storage

0.09

 Waiting

0.09

Activations Density 0.127%