INDEX

Explanations

aggressive use of mathematical symbols and references to quantifiable data

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

iph

-0.08

erville

-0.07

sequential

-0.07

Ã©ru

-0.07

getError

-0.07

otec

-0.07

esan

-0.06

angstrom

-0.06

ÑĢÐ¾Ñĩ

-0.06

POSITIVE LOGITS

ch

0.07

 Macro

0.07

 macro

0.07

Obr

0.06

macro

0.06

/mac

0.06

MAC

0.06

Macro

0.06

imes

0.06

 interaction

0.06

Activations Density 0.001%