INDEX

Explanations

specifies or indicates conditions

The neuron identifies imperative overview phrases in code or API documentation comments (e.g. “Initializes…,” “Specifies…,” “Occurs when…”) that introduce summary descriptions.

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

 impossível

-1.63

ܕ

-1.61

 černé

-1.59

 halben

-1.54

 hlavně

-1.52

 barvy

-1.49

鶫

-1.44

 traduc

-1.43

 joyful

-1.42

ޏ

-1.42

POSITIVE LOGITS

if

2.47

is

2.17

or

2.13

 indicates

1.67

 used

1.66

has

1.63

an

1.59

the

1.52

there

1.52

as

1.50

Activations Density 0.047%