INDEX

Explanations

punctuation and conversational cues in dialogue

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

elper

-0.07

ampa

-0.06

arel

-0.06

_guide

-0.06

ëĭ¤ê°Ģ

-0.06

 Brew

-0.06

eydi

-0.06

insk

-0.06

ayi

-0.06

åĲ¹

-0.06

POSITIVE LOGITS

 Dire

0.06

 salle

0.06

igos

0.06

akov

0.06

Ð°Ð½Ðº

0.06

Ð¸Ð½Ð¾Ð²

0.06

udo

0.06

aku

0.06

 Weinstein

0.06

PUTE

0.06

Activations Density 0.005%