INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 challenger

-0.11

 adul

-0.11

ieran

-0.10

ernel

-0.10

ensch

-0.10

 Mast

-0.10

 budding

-0.10

 Smoking

-0.10

 Reconstruction

-0.10

 buffered

-0.10

POSITIVE LOGITS

 alike

0.09

-fly

0.08

 stigma

0.08

 unaware

0.08

sealed

0.08

-radius

0.07

iki

0.07

UID

0.07

 veto

0.07

CHR

0.07

Activations Density 0.000%

No Known Activations