Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 85 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0331. Min Act: -0.0005

Data Index: 644752 (C4 (Web Text))

Max Activating Token Index: 329

Text #1

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0338. Min Act: -0.0005

Data Index: 197204 (C4 (Web Text))

Max Activating Token Index: 532

Text #2

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0294. Min Act: -0.0005

Data Index: 593511 (C4 (Web Text))

Max Activating Token Index: 375

Text #3

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0260. Min Act: -0.0005

Data Index: 694675 (C4 (Web Text))

Max Activating Token Index: 426

Text #4

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0288. Min Act: -0.0005

Data Index: 12077 (C4 (Web Text))

Max Activating Token Index: 967

Text #5

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0284. Min Act: -0.0005

Data Index: 459684 (C4 (Web Text))

Max Activating Token Index: 959

Text #6

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0271. Min Act: -0.0005

Data Index: 713790 (C4 (Web Text))

Max Activating Token Index: 139

Text #7

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0268. Min Act: -0.0005

Data Index: 475598 (C4 (Web Text))

Max Activating Token Index: 956

Text #8

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0264. Min Act: -0.0005

Data Index: 389194 (C4 (Web Text))

Max Activating Token Index: 647

Text #9

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0265. Min Act: -0.0005

Data Index: 712133 (C4 (Web Text))

Max Activating Token Index: 661

Text #10

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0261. Min Act: -0.0005

Data Index: 793915 (C4 (Web Text))

Max Activating Token Index: 207

Text #11

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0277. Min Act: -0.0005

Data Index: 778440 (C4 (Web Text))

Max Activating Token Index: 840

Text #12

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0255. Min Act: -0.0005

Data Index: 398470 (C4 (Web Text))

Max Activating Token Index: 860

Text #13

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0260. Min Act: -0.0005

Data Index: 1256531 (C4 (Web Text))

Max Activating Token Index: 305

Text #14

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0268. Min Act: -0.0005

Data Index: 1356729 (C4 (Web Text))

Max Activating Token Index: 596

Text #15

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0272. Min Act: -0.0005

Data Index: 306640 (C4 (Web Text))

Max Activating Token Index: 317

Text #16

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0267. Min Act: -0.0005

Data Index: 904404 (C4 (Web Text))

Max Activating Token Index: 179

Text #17

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0264. Min Act: -0.0005

Data Index: 170147 (C4 (Web Text))

Max Activating Token Index: 646

Text #18

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0269. Min Act: -0.0005

Data Index: 15134 (C4 (Web Text))

Max Activating Token Index: 560

Text #19

Max Range: 0.0338. Min Range: -0.0338

Max Act: 0.0246. Min Act: -0.0005

Data Index: 393107 (C4 (Web Text))

Max Activating Token Index: 218