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

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

Neuron 5933 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.7540. Min Act: -0.0000

Data Index: 465198 (C4 (Web Text))

Max Activating Token Index: 516

Text #1

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.7444. Min Act: -0.0000

Data Index: 218978 (C4 (Web Text))

Max Activating Token Index: 256

Text #2

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.6382. Min Act: -0.0000

Data Index: 536342 (C4 (Web Text))

Max Activating Token Index: 997

Text #3

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.6405. Min Act: -0.0000

Data Index: 530596 (C4 (Web Text))

Max Activating Token Index: 609

Text #4

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.6516. Min Act: -0.0000

Data Index: 373627 (C4 (Web Text))

Max Activating Token Index: 377

Text #5

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.6111. Min Act: -0.0000

Data Index: 642538 (C4 (Web Text))

Max Activating Token Index: 833

Text #6

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.5927. Min Act: -0.0000

Data Index: 170704 (C4 (Web Text))

Max Activating Token Index: 548

Text #7

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.6055. Min Act: -0.0000

Data Index: 672572 (C4 (Web Text))

Max Activating Token Index: 847

Text #8

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.5306. Min Act: -0.0000

Data Index: 816311 (C4 (Web Text))

Max Activating Token Index: 302

Text #9

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.5416. Min Act: -0.0000

Data Index: 357300 (C4 (Web Text))

Max Activating Token Index: 832

Text #10

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.5073. Min Act: -0.0000

Data Index: 885063 (C4 (Web Text))

Max Activating Token Index: 301

Text #11

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.5377. Min Act: -0.0000

Data Index: 21028 (C4 (Web Text))

Max Activating Token Index: 614

Text #12

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.5106. Min Act: -0.0000

Data Index: 756328 (C4 (Web Text))

Max Activating Token Index: 720

Text #13

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.4927. Min Act: -0.0000

Data Index: 606742 (C4 (Web Text))

Max Activating Token Index: 653

Text #14

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.4874. Min Act: -0.0000

Data Index: 378339 (C4 (Web Text))

Max Activating Token Index: 295

Text #15

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.5168. Min Act: -0.0000

Data Index: 246230 (C4 (Web Text))

Max Activating Token Index: 1015

Text #16

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.4680. Min Act: -0.0000

Data Index: 462507 (C4 (Web Text))

Max Activating Token Index: 854

Text #17

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.4537. Min Act: -0.0000

Data Index: 653664 (C4 (Web Text))

Max Activating Token Index: 242

Text #18

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.4805. Min Act: -0.0000

Data Index: 1585261 (Python Code)

Max Activating Token Index: 552

Text #19

Max Range: 0.7540. Min Range: -0.7540

Max Act: 0.4618. Min Act: -0.0000

Data Index: 967199 (C4 (Web Text))

Max Activating Token Index: 548