INDEX

Explanations

discussions about the limits of human understanding and consciousness

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

VRTX

-0.07

elper

-0.07

implicit

-0.07

reuse

-0.07

alette

-0.07

implicitly

-0.07

stile

-0.07

keit

-0.07

orie

-0.07

ecycle

-0.07

POSITIVE LOGITS

 comprehension

0.07

 limit

0.07

 limited

0.07

 mystery

0.06

 Calder

0.06

Cop

0.06

 Cant

0.06

 human

0.06

nder

0.06

Cop

0.06

Activations Density 0.024%