INDEX

Explanations

The neuron strongly activates on mentions of “power” (and related political or authority jargon).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 cDNA

0.63

 lindos

0.63

 keepsake

0.63

 استقبال

0.62

 શુભેચ્છ

0.61

鲷

0.61

 biodegrad

0.61

letscher

0.60

छात्रा

0.59

 biodegradation

0.59

POSITIVE LOGITS

 power

2.64

power

2.36

 potere

2.20

 POWER

2.19

 Power

2.17

Power

2.13

权力

2.09

POWER

2.02

 पावर

1.97

 власть

1.95

Activations Density 0.986%