INDEX

Explanations

worker processes

The neuron fires on technical labels for system “nodes” or roles—specifically on terms like “leaf” (as in leaf nodes) and “worker(s)” (as in worker processes).

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

ти

-1.62

Δ

-1.41

綃

-1.38

ší

-1.37

accepted

-1.33

 Институ

-1.32

{},

-1.30

逦

-1.28

 integran

-1.27

POSITIVE LOGITS

 моя

1.73

but

1.72

but

1.64

 extraordinaire

1.60

 variasi

1.59

 vínculo

1.57

 paraíso

1.55

izes

1.54

 afectado

1.52

 moje

1.51

Activations Density 0.023%