INDEX

Explanations

ranges, sizes, durations, frequencies

The neuron detects explicit numerical or quantitative terms (numbers, measurements, units) in technical or data‐heavy contexts.

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

 after

-1.66

 only

-1.59

 would

-1.30

two

-1.26

 just

-1.25

 without

-1.24

 before

-1.24

 took

-1.22

 first

-1.22

 when

-1.22

POSITIVE LOGITS

 ranging

3.09

ranging

2.70

 ranges

1.88

 ranged

1.84

anging

1.83

 range

1.72

 varier

1.70

 mulai

1.70

 allant

1.64

 pouvant

1.60

Activations Density 0.205%