INDEX

Explanations

handling requests

The neuron consistently activates on uses of the HTTP request object (tokens like `req`, `request`, or `.get` on those identifiers).

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

of

-2.02

我是

-1.52

 alltid

-1.47

 vždy

-1.44

𖧵

-1.43

 всегда

-1.42

 '\\;'

-1.42

 anderen

-1.41

 always

-1.39

 altijd

-1.38

POSITIVE LOGITS

没能

1.83

並沒有

1.72

无比

1.62

这般

1.59

进入了

1.57

并未

1.56

面對

1.55

 dourado

1.52

给我们

1.52

ebaran

1.51

Activations Density 0.001%