INDEX

Explanations

same followed by context

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 Garantía

-1.29

 vemos

-1.26

место

-1.23

 после

-1.19

लों

-1.18

still

-1.16

 toujours

-1.16

部分

-1.14

 still

-1.13

 począ

-1.13

POSITIVE LOGITS

 with

1.77

for

1.61

čení

1.38

1.20

chenden

1.16

 these

1.15

钢琴

1.14

setToolTipText

1.13

even

1.12

 выгод

1.09

Activations Density 0.061%

No Known Activations