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

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

Neuron 110 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: 2.6523. Min Range: -2.6523

Max Act: 2.6523. Min Act: -0.0000

Data Index: 579722 (C4 (Web Text))

Max Activating Token Index: 714

Text #1

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.6265. Min Act: -0.0000

Data Index: 458962 (C4 (Web Text))

Max Activating Token Index: 460

Text #2

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.5843. Min Act: -0.0000

Data Index: 157963 (C4 (Web Text))

Max Activating Token Index: 783

Text #3

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.5319. Min Act: -0.0000

Data Index: 725515 (C4 (Web Text))

Max Activating Token Index: 199

Text #4

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.5302. Min Act: -0.0000

Data Index: 52638 (C4 (Web Text))

Max Activating Token Index: 152

Text #5

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.4937. Min Act: -0.0000

Data Index: 376984 (C4 (Web Text))

Max Activating Token Index: 378

Text #6

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.4882. Min Act: -0.0000

Data Index: 601733 (C4 (Web Text))

Max Activating Token Index: 668

Text #7

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.4809. Min Act: -0.0000

Data Index: 583473 (C4 (Web Text))

Max Activating Token Index: 446

Text #8

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.4516. Min Act: -0.0000

Data Index: 1089020 (C4 (Web Text))

Max Activating Token Index: 90

Text #9

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.4474. Min Act: -0.0000

Data Index: 1058048 (C4 (Web Text))

Max Activating Token Index: 108

Text #10

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.4388. Min Act: -0.0000

Data Index: 471069 (C4 (Web Text))

Max Activating Token Index: 994

Text #11

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.2860. Min Act: -0.0000

Data Index: 318800 (C4 (Web Text))

Max Activating Token Index: 176

Text #12

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.5355. Min Act: -0.0000

Data Index: 492215 (C4 (Web Text))

Max Activating Token Index: 115

Text #13

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.4038. Min Act: -0.0000

Data Index: 1107461 (C4 (Web Text))

Max Activating Token Index: 370

Text #14

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.4597. Min Act: -0.0000

Data Index: 1432725 (Python Code)

Max Activating Token Index: 255

Text #15

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.1831. Min Act: -0.0000

Data Index: 1073981 (C4 (Web Text))

Max Activating Token Index: 770

Text #16

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.3050. Min Act: -0.0000

Data Index: 1082725 (C4 (Web Text))

Max Activating Token Index: 76

Text #17

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.3134. Min Act: -0.0000

Data Index: 303794 (C4 (Web Text))

Max Activating Token Index: 668

Text #18

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.3929. Min Act: -0.0000

Data Index: 446800 (C4 (Web Text))

Max Activating Token Index: 918

Text #19

Max Range: 2.6523. Min Range: -2.6523

Max Act: 2.1670. Min Act: -0.0000

Data Index: 179963 (C4 (Web Text))

Max Activating Token Index: 296