INDEX

Explanations

references to prior academic work and studies

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

orm

-0.07

apr

-0.06

ì´

-0.06

UpdatedAt

-0.06

 Wheeler

-0.06

 wikipedia

-0.06

ï»¿using

-0.06

Anth

-0.06

ORM

-0.06

 Wikipedia

-0.06

POSITIVE LOGITS

 previously

0.08

 Previously

0.07

 paper

0.07

Previously

0.07

_paper

0.07

me

0.07

 myself

0.07

ĥĿ

0.07

 previous

0.06

 Paper

0.06

Activations Density 0.075%