INDEX

Explanations

filter function

The neuron strongly activates on occurrences of “Filter” (e.g. class names, log entries, or config keys referring to filters) in code or configuration text.

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

ване

-0.85

 pyrite

-0.83

 địch

-0.78

세요

-0.77

月に

-0.76

文学

-0.74

 benzo

-0.74

गा

-0.73

 Trucking

-0.72

UDA

-0.70

POSITIVE LOGITS

 filter

1.61

 Filter

1.55

 filters

1.39

filter

1.33

Filter

1.30

setFilter

1.29

过滤器

1.22

 filtra

1.21

filters

1.20

 Filters

1.13

Activations Density 0.005%