INDEX

Explanations

phrases indicating potential outcomes or results of actions

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

ecta

-0.06

erule

-0.06

nor

-0.06

ims

-0.06

ICA

-0.06

yor

-0.06

veloper

-0.05

continental

-0.05

ALA

-0.05

alar

-0.05

POSITIVE LOGITS

 fruition

0.17

 pass

0.15

 fruit

0.13

 completion

0.12

 reality

0.12

fr

0.11

fr

0.11

 fruits

0.11

 light

0.11

 realization

0.11

Activations Density 0.008%