INDEX

Explanations

conjunctions and related linguistic structures that contribute to sentence flow and complexity

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

ead

-0.07

trs

-0.07

LOY

-0.07

aria

-0.07

 suppress

-0.07

ivery

-0.07

eka

-0.06

abb

-0.06

.slim

-0.06

Ã¡h

-0.06

POSITIVE LOGITS

 being

0.14

 having

0.10

 getting

0.10

being

0.10

 becoming

0.09

 making

0.08

 Being

0.08

ÏİÎ½ÏĦÎ±ÏĤ

0.08

 correl

0.08

 doing

0.08

Activations Density 0.328%