INDEX

Explanations

months or years duration

The neuron detects references to armed conflict or military contexts (e.g., war zones, battalions, soldiers).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

并且

0.85

 Međutim

0.82

 પરંતુ

0.80

üglich

0.77

ധികം

0.77

 Piccolo

0.77

 snorkeling

0.77

 ngunit

0.76

そして

0.75

এবং

0.74

POSITIVE LOGITS

或

2.64

or

2.52

หรือ

2.39

 или

2.38

 atau

2.38

 veya

2.30

或是

2.25

 hoặc

2.22

 یا

2.13

或

2.11

Activations Density 3.254%