INDEX

Explanations

online platforms and services

The neuron detects mentions of website or online platform names (e.g., Twitter, YouTube, Amazon.com).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ortheast

-1.29

₈

-1.25

%",

-1.20

)$.

-1.20

 either

-1.19

’;

-1.18

?’

-1.16

»).

-1.16

CDM

-1.15

沄

-1.15

POSITIVE LOGITS

1.77

 όλα

1.32

 better

1.27

 September

1.27

 October

1.21

 todos

1.19

 wszystko

1.18

 December

1.17

Activations Density 0.018%