INDEX

Explanations

expressions and discussions surrounding truth and reality

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

itle

-0.07

curity

-0.06

plr

-0.06

configs

-0.06

utters

-0.06

_RESOLUTION

-0.06

 ÑĢÐ°Ð²Ð½Ð¾

-0.06

 palate

-0.06

 merk

-0.06

reet

-0.06

POSITIVE LOGITS

 versus

0.08

usra

0.07

una

0.07

-REAL

0.07

 whether

0.07

çľŁæĺ¯

0.06

vers

0.06

avou

0.06

 exactly

0.06

ãĤ¹ãĥ¬

0.06

Activations Density 0.066%