INDEX

Explanations

academic journal abbreviations

The neuron primarily detects academic journal titles or their standard abbreviations in reference citations.

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

-1.90

讥

-1.69

For

-1.51

 ‘‘

-1.39

an

-1.38

-1.37

“

-1.36

脾气

-1.32

-1.30

 after

-1.29

POSITIVE LOGITS

of

2.44

 bénévoles

1.91

ｎ

1.87

 lunghi

1.73

这种

1.71

厹

1.71

 orologio

1.70

 naszych

1.69

 occhiali

1.69

 této

1.68

Activations Density 0.002%