occurrences and actions

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 Then

-1.23

 desde

-1.20

 jedin

-1.16

 takich

-1.15

 disregard

-1.13

 Vordergrund

-1.13

 That

-1.13

 Это

-1.12

 désir

-1.12

 dipas

-1.11

POSITIVE LOGITS

 piac

1.24

 Sticks

1.21

 immerhin

1.20

 pollut

1.20

 possíveis

1.20

榫

1.19

 diser

1.17

 labi

1.17

是否是

1.16

Activations Density 0.088%

No Known Activations