INDEX

Explanations

curly braces and return statements in code

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

hood

-0.07

burn

-0.06

arger

-0.06

hs

-0.06

ester

-0.06

jl

-0.06

quared

-0.06

 types

-0.06

lace

-0.06

POSITIVE LOGITS

atk

0.08

 Atlas

0.07

ixmap

0.07

 Ð¿Ð¾Ð±Ð°

0.07

Atlas

0.07

asher

0.07

"value

0.07

 INLINE

0.07

vanced

0.07

 Ð³Ð¾Ð»Ð¾Ð²Ð¸

0.07

Activations Density 0.003%