INDEX

Explanations

free or exempt tax/fees

The neuron activates on occurrences of “free” (including in compounds like “toll-free,” “royalty-free,” etc.).

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

笮

-2.02

そして

-1.95

 και

-1.92

«

-1.74

</h6>

-1.71

そういえば

-1.67

逐渐

-1.65

-1.63

ቧ

-1.61

POSITIVE LOGITS

in

1.79

at

1.71

jauh

1.59

but

1.59

の神

1.58

employee

1.57

 Início

1.56

 llevó

1.55

lishes

1.53

 acudir

1.45

Activations Density 0.003%