INDEX

Explanations

Names and dates

The neuron detects HTML table-cell elements (i.e. <td>…</td> tags).

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

-1.84

</h1>

-1.55

羕

-1.52

â

-1.43

but

-1.38

“

-1.37

ironically

-1.32

 Dingen

-1.29

だけでなく

-1.28

evident

-1.27

POSITIVE LOGITS

 problemet

1.73

 francisco

1.73

 bidra

1.59

ஂ

1.58

 peruano

1.55

mtu

1.54

muting

1.53

 démocr

1.51

 espirituales

1.51

phu

1.50

Activations Density 0.000%