INDEX

Explanations

html input attributes

The neuron detects HTML “input” elements (i.e. the presence of an `<input>` tag with its type attribute).

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

-2.19

os

-1.84

 Then

-1.84

 Mainly

-1.73

 According

-1.70

起来的

-1.64

那时

-1.62

um

-1.59

躋

-1.58

who

-1.57

POSITIVE LOGITS

 skrive

1.88

damas

1.86

 sikre

1.83

 høre

1.79

ſelves

1.77

 konflikt

1.76

and

1.74

šla

1.72

 Дмитрий

1.70

år

1.68

Activations Density 0.027%