INDEX

Explanations

monthly archives

This neuron detects HTML markup elements, especially link tags and related attributes.

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

 usually

-0.89

six

-0.88

 should

-0.86

 required

-0.84

方がいい

-0.82

 unusual

-0.82

始まる

-0.82

 often

-0.82

 pushed

-0.81

 enormous

-0.81

POSITIVE LOGITS

Traversal

1.06

ViewHolder

0.99

SMC

0.99

ніе

0.97

édez

0.95

 vécu

0.94



0.93

VIAF

0.93

zież

0.93

ctis

0.93

Activations Density 0.007%