INDEX

Explanations

code-specific outputs, error messages, and data result assignments

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

EMS

-0.07

sty

-0.07

 eskort

-0.07

endi

-0.07

eson

-0.07

æĺŁæľŁ

-0.07

cÃŃ

-0.06

íĨłíĨł

-0.06

dden

-0.06

rani

-0.06

POSITIVE LOGITS

 Goodman

0.06

execute

0.06

||||

0.06

 ÑĥÑģ

0.06

erek

0.05

Jar

0.05

RICT

0.05

jar

0.05

ë¬¼

0.05

.execute

0.05

Activations Density 0.029%