INDEX

Explanations

subvert, version, subversion

This neuron detects occurrences of the root “subver” (as in subvert, subversion).

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

猻

-2.50

躥

-2.34

៚

-2.33

 climatisation

-2.09

廰

-2.03

iteľ

-2.02

お金

-1.99

 fantástica

-1.98

黻

-1.97

 revisor

-1.96

POSITIVE LOGITS

↵

2.64

他

2.47

}$

2.44

<i>

2.36

</b>

2.23

这一

2.20

<b>

2.14

媜

2.09

 Thanks

2.05



2.05

Activations Density 0.004%