INDEX

Explanations

rates and pricing

The neuron detects mentions of pricing information—especially numbers or words like “rate(s)” associated with costs.

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

㩒

-1.81

 habilidades

-1.74

 handeln

-1.50

キリスト

-1.48

何故

-1.48

骐

-1.48

 Nearly

-1.47

<bos>

-1.46

그리고

-1.43

ﾑ

-1.42

POSITIVE LOGITS

 THOSE

1.50

up

1.50

to

1.49

 bruin

1.45

?"

1.36

اتها

1.31

lo

1.30

 tien

1.30

 oogen

1.30

Activations Density 0.023%