INDEX

Explanations

phrases indicating assumptions and knowledge claims

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

_NOW

-0.08

 NavParams

-0.08

roj

-0.08

Cob

-0.07

å²

-0.07

unce

-0.07

zan

-0.07

agu

-0.07

_Lean

-0.07

enties

-0.07

POSITIVE LOGITS

 premise

0.07

ãĥĭãĤ¢

0.06

.want

0.06

 assumption

0.06

 assumed

0.06

 assuming

0.05

 that

0.05

 assumptions

0.05

want

0.05

Activations Density 0.029%