Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1612 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0400. Min Act: -0.0003

Data Index: 1391631 (Python Code)

Max Activating Token Index: 1023

Text #1

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0385. Min Act: -0.0003

Data Index: 962702 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0381. Min Act: -0.0003

Data Index: 287997 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0391. Min Act: -0.0003

Data Index: 1026857 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0382. Min Act: -0.0003

Data Index: 394581 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0371. Min Act: -0.0003

Data Index: 1650510 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0371. Min Act: -0.0003

Data Index: 575347 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0351. Min Act: -0.0003

Data Index: 305482 (C4 (Web Text))

Max Activating Token Index: 847

Text #8

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0349. Min Act: -0.0003

Data Index: 892216 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0360. Min Act: -0.0002

Data Index: 1530198 (Python Code)

Max Activating Token Index: 1023

Text #10

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0316. Min Act: -0.0003

Data Index: 1201731 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0357. Min Act: -0.0004

Data Index: 252571 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0350. Min Act: -0.0003

Data Index: 710983 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0362. Min Act: -0.0003

Data Index: 414123 (C4 (Web Text))

Max Activating Token Index: 1019

Text #14

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0344. Min Act: -0.0004

Data Index: 1050613 (C4 (Web Text))

Max Activating Token Index: 268

Text #15

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0349. Min Act: -0.0003

Data Index: 828144 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0344. Min Act: -0.0002

Data Index: 778271 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0347. Min Act: -0.0003

Data Index: 1528263 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0349. Min Act: -0.0003

Data Index: 575895 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 0.0400. Min Range: -0.0400

Max Act: 0.0345. Min Act: -0.0003

Data Index: 307760 (C4 (Web Text))

Max Activating Token Index: 1023