Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1908 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 2.7830. Min Range: -2.7830

Max Act: 2.6535. Min Act: -0.0000

Data Index: 1183167 (C4 (Web Text))

Max Activating Token Index: 486

Text #1

Max Range: 2.7830. Min Range: -2.7830

Max Act: 2.7830. Min Act: -0.0001

Data Index: 928307 (C4 (Web Text))

Max Activating Token Index: 326

Text #2

Max Range: 2.7830. Min Range: -2.7830

Max Act: 2.0865. Min Act: -0.0000

Data Index: 86413 (C4 (Web Text))

Max Activating Token Index: 337

Text #3

Max Range: 2.7830. Min Range: -2.7830

Max Act: 2.0504. Min Act: -0.0000

Data Index: 1470714 (Python Code)

Max Activating Token Index: 431

Text #4

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.9669. Min Act: -0.0001

Data Index: 1587884 (Python Code)

Max Activating Token Index: 250

Text #5

Max Range: 2.7830. Min Range: -2.7830

Max Act: 2.0732. Min Act: -0.0000

Data Index: 1123193 (C4 (Web Text))

Max Activating Token Index: 739

Text #6

Max Range: 2.7830. Min Range: -2.7830

Max Act: 2.0543. Min Act: -0.0000

Data Index: 1411956 (Python Code)

Max Activating Token Index: 203

Text #7

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.8990. Min Act: -0.0000

Data Index: 1206695 (C4 (Web Text))

Max Activating Token Index: 601

Text #8

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.9939. Min Act: -0.0000

Data Index: 228604 (C4 (Web Text))

Max Activating Token Index: 888

Text #9

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.9212. Min Act: -0.0000

Data Index: 1613365 (Python Code)

Max Activating Token Index: 189

Text #10

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.8904. Min Act: -0.0000

Data Index: 868550 (C4 (Web Text))

Max Activating Token Index: 362

Text #11

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.8105. Min Act: -0.0001

Data Index: 178219 (C4 (Web Text))

Max Activating Token Index: 285

Text #12

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.8343. Min Act: -0.0001

Data Index: 799542 (C4 (Web Text))

Max Activating Token Index: 446

Text #13

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.7383. Min Act: -0.0000

Data Index: 1031320 (C4 (Web Text))

Max Activating Token Index: 692

Text #14

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.7351. Min Act: -0.0001

Data Index: 1476966 (Python Code)

Max Activating Token Index: 428

Text #15

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.8250. Min Act: -0.0001

Data Index: 928766 (C4 (Web Text))

Max Activating Token Index: 1009

Text #16

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.7754. Min Act: -0.0000

Data Index: 1648311 (Python Code)

Max Activating Token Index: 981

Text #17

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.5745. Min Act: -0.0001

Data Index: 833517 (C4 (Web Text))

Max Activating Token Index: 765

Text #18

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.5779. Min Act: -0.0001

Data Index: 1311305 (C4 (Web Text))

Max Activating Token Index: 950

Text #19

Max Range: 2.7830. Min Range: -2.7830

Max Act: 1.7046. Min Act: -0.0000

Data Index: 1314330 (C4 (Web Text))

Max Activating Token Index: 726