INDEX

Explanations

the word "como" and its variations, indicating comparisons

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Negative Logits

ses

-0.08

ors

-0.08

acades

-0.07

igner

-0.07

roe

-0.07

 kalan

-0.07

opleft

-0.07

·»

-0.07

anvas

-0.07

edImage

-0.07

POSITIVE LOGITS

 follows

0.08

cert

0.08

Å¾

0.08

Ð¸Ð¼Ð¸

0.07

-ÑĤÐ¾

0.07

oci

0.06

pires

0.06

icone

0.06

 sendo

0.06

 Å¡to

0.06

Activations Density 0.009%

No Known Activations