INDEX

Explanations

questions, forums, names, dates

The neuron detects metadata timestamps and similar header information (dates/times in document headers).

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

 мужская

-1.92

$\

-1.73

mathrm

-1.67

 женская

-1.62

 figlia

-1.60

:"#

-1.57

 gemeenten

-1.57

()){

-1.56

/////////////

-1.55

-1.54

POSITIVE LOGITS

 abusi

1.42

 sinu

1.41

1.39

 stale

1.38

塒

1.37

を見ると

1.36

 BITS

1.35

 clichés

1.33

LIBRA

1.32

Можно

1.31

Activations Density 0.000%