INDEX

Explanations

energy infrastructure and transportation

The main thing this neuron does is detect mentions of pipelines or pipeline‐related project names.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

-1.27

 vœux

-1.21

Heaven

-1.17

–

-1.16

abbix

-1.14

 lẽ

-1.13

 wenigstens

-1.11

<blockquote>

-1.11

-1.10

lanted

-1.09

POSITIVE LOGITS



1.37

zja

1.34

how

1.26

 vacía

1.25

 capitulo

1.24

urally

1.24

 possa

1.23

᠁

1.21

})$}

1.18

瓛

1.17

Activations Density 0.058%