INDEX

Explanations

phrases indicating adaptation and customization for specific audiences or contexts

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Embeds

IFrame

Link

Negative Logits

echan

-0.08

zel

-0.07

anto

-0.07

bose

-0.07

ampo

-0.06

overview

-0.06

rosso

-0.06

anship

-0.06

_WAKE

-0.06

bay

-0.06

POSITIVE LOGITS

 needs

0.12

 situation

0.11

 circumstances

0.11

 realities

0.10

 context

0.10

 unique

0.10

 circumstance

0.10

 situations

0.10

needs

0.10

 specific

0.10

Activations Density 0.047%

No Known Activations