Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 3305 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.4806. Min Range: -2.4806

Max Act: 2.4806. Min Act: -0.0000

Data Index: 87148 (C4 (Web Text))

Max Activating Token Index: 264

Text #1

Max Range: 2.4806. Min Range: -2.4806

Max Act: 2.2548. Min Act: -0.0000

Data Index: 799413 (C4 (Web Text))

Max Activating Token Index: 840

Text #2

Max Range: 2.4806. Min Range: -2.4806

Max Act: 2.1006. Min Act: -0.0000

Data Index: 1200698 (C4 (Web Text))

Max Activating Token Index: 219

Text #3

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.9657. Min Act: -0.0000

Data Index: 22865 (C4 (Web Text))

Max Activating Token Index: 219

Text #4

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.9355. Min Act: -0.0000

Data Index: 491518 (C4 (Web Text))

Max Activating Token Index: 551

Text #5

Max Range: 2.4806. Min Range: -2.4806

Max Act: 2.0067. Min Act: -0.0000

Data Index: 1201223 (C4 (Web Text))

Max Activating Token Index: 117

Text #6

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.9845. Min Act: -0.0000

Data Index: 1243493 (C4 (Web Text))

Max Activating Token Index: 16

Text #7

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.9631. Min Act: -0.0000

Data Index: 288706 (C4 (Web Text))

Max Activating Token Index: 680

Text #8

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.8853. Min Act: -0.0000

Data Index: 524047 (C4 (Web Text))

Max Activating Token Index: 578

Text #9

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.8516. Min Act: -0.0000

Data Index: 979883 (C4 (Web Text))

Max Activating Token Index: 921

Text #10

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.7402. Min Act: -0.0000

Data Index: 886544 (C4 (Web Text))

Max Activating Token Index: 271

Text #11

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.7233. Min Act: -0.0000

Data Index: 659539 (C4 (Web Text))

Max Activating Token Index: 729

Text #12

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.6677. Min Act: -0.0000

Data Index: 59829 (C4 (Web Text))

Max Activating Token Index: 381

Text #13

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.6676. Min Act: -0.0000

Data Index: 836679 (C4 (Web Text))

Max Activating Token Index: 500

Text #14

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.4045. Min Act: -0.0000

Data Index: 1621218 (Python Code)

Max Activating Token Index: 729

Text #15

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.5253. Min Act: -0.0000

Data Index: 1127799 (C4 (Web Text))

Max Activating Token Index: 102

Text #16

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.5280. Min Act: -0.0000

Data Index: 1338478 (C4 (Web Text))

Max Activating Token Index: 574

Text #17

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.6907. Min Act: -0.0000

Data Index: 333525 (C4 (Web Text))

Max Activating Token Index: 696

Text #18

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.5665. Min Act: -0.0000

Data Index: 284477 (C4 (Web Text))

Max Activating Token Index: 771

Text #19

Max Range: 2.4806. Min Range: -2.4806

Max Act: 1.5457. Min Act: -0.0000

Data Index: 373276 (C4 (Web Text))

Max Activating Token Index: 837