INDEX

Explanations

references to collaborative efforts and actions that involve groups or organizations

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

Ð½Ð°Ð´Ð»ÐµÐ¶

-0.08

eko

-0.07

loom

-0.06

voir

-0.06

rchive

-0.06

 blown

-0.06

amp

-0.06

 Berry

-0.06

ÌĨ

-0.06

POSITIVE LOGITS

 itself

0.08

 involved

0.07

is

0.07

 involves

0.07

's

0.07

obot

0.06

ckt

0.06

 algÃºn

0.06

 concerned

0.06

 still

0.06

Activations Density 0.039%