Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4522 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.0102. Min Range: -1.0102

Max Act: 1.0102. Min Act: -0.0000

Data Index: 836739 (C4 (Web Text))

Max Activating Token Index: 998

Text #1

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.9697. Min Act: -0.0000

Data Index: 1387010 (Python Code)

Max Activating Token Index: 20

Text #2

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.9668. Min Act: -0.0001

Data Index: 428162 (C4 (Web Text))

Max Activating Token Index: 972

Text #3

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.8407. Min Act: -0.0001

Data Index: 279573 (C4 (Web Text))

Max Activating Token Index: 880

Text #4

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.8610. Min Act: -0.0000

Data Index: 259934 (C4 (Web Text))

Max Activating Token Index: 279

Text #5

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.8313. Min Act: -0.0000

Data Index: 778784 (C4 (Web Text))

Max Activating Token Index: 57

Text #6

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.8321. Min Act: -0.0000

Data Index: 1498521 (Python Code)

Max Activating Token Index: 854

Text #7

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.8304. Min Act: -0.0001

Data Index: 763657 (C4 (Web Text))

Max Activating Token Index: 425

Text #8

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.8195. Min Act: -0.0001

Data Index: 235254 (C4 (Web Text))

Max Activating Token Index: 272

Text #9

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.7729. Min Act: -0.0001

Data Index: 1585163 (Python Code)

Max Activating Token Index: 93

Text #10

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.8182. Min Act: -0.0001

Data Index: 433192 (C4 (Web Text))

Max Activating Token Index: 964

Text #11

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.8032. Min Act: -0.0001

Data Index: 828400 (C4 (Web Text))

Max Activating Token Index: 442

Text #12

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.7599. Min Act: -0.0001

Data Index: 285148 (C4 (Web Text))

Max Activating Token Index: 545

Text #13

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.8365. Min Act: -0.0000

Data Index: 1149664 (C4 (Web Text))

Max Activating Token Index: 189

Text #14

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.7666. Min Act: -0.0001

Data Index: 1028664 (C4 (Web Text))

Max Activating Token Index: 204

Text #15

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.7995. Min Act: -0.0001

Data Index: 225118 (C4 (Web Text))

Max Activating Token Index: 98

Text #16

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.6975. Min Act: -0.0000

Data Index: 691075 (C4 (Web Text))

Max Activating Token Index: 60

Text #17

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.6980. Min Act: -0.0001

Data Index: 1420679 (Python Code)

Max Activating Token Index: 963

Text #18

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.7642. Min Act: -0.0001

Data Index: 112043 (C4 (Web Text))

Max Activating Token Index: 491

Text #19

Max Range: 1.0102. Min Range: -1.0102

Max Act: 0.7614. Min Act: -0.0001

Data Index: 914210 (C4 (Web Text))

Max Activating Token Index: 603