INDEX

Explanations

occurrences of the word "with" in various contexts

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

ê°ľë¥¼

-0.08

ÏħÏĦÏĮ

-0.08

ãĥ©ãĤ¤ãĥĪ

-0.08

ä¹İ

-0.08

 ë¬´ìĹĩ

-0.08

ÐłÐĿ

-0.08

plevel

-0.08

avit

-0.08

vfs

-0.07

iterals

-0.07

POSITIVE LOGITS

 whom

0.11

 whose

0.08

who

0.07

 disabilities

0.07

 access

0.07

 valid

0.07

stood

0.06

 similar

0.06

whose

0.06

 experience

0.06

Activations Density 0.020%