Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 998 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.6447. Min Act: -0.0002

Data Index: 261670 (C4 (Web Text))

Max Activating Token Index: 369

Text #1

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.5414. Min Act: -0.0002

Data Index: 825765 (C4 (Web Text))

Max Activating Token Index: 556

Text #2

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.5092. Min Act: -0.0002

Data Index: 903960 (C4 (Web Text))

Max Activating Token Index: 982

Text #3

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.5077. Min Act: -0.0002

Data Index: 195339 (C4 (Web Text))

Max Activating Token Index: 820

Text #4

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.5077. Min Act: -0.0001

Data Index: 1548853 (Python Code)

Max Activating Token Index: 698

Text #5

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.5115. Min Act: -0.0001

Data Index: 164958 (C4 (Web Text))

Max Activating Token Index: 118

Text #6

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.5201. Min Act: -0.0002

Data Index: 205099 (C4 (Web Text))

Max Activating Token Index: 546

Text #7

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4130. Min Act: -0.0002

Data Index: 305749 (C4 (Web Text))

Max Activating Token Index: 362

Text #8

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4410. Min Act: -0.0002

Data Index: 999004 (C4 (Web Text))

Max Activating Token Index: 947

Text #9

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4685. Min Act: -0.0002

Data Index: 888569 (C4 (Web Text))

Max Activating Token Index: 364

Text #10

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4490. Min Act: -0.0002

Data Index: 1253433 (C4 (Web Text))

Max Activating Token Index: 11

Text #11

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4350. Min Act: -0.0001

Data Index: 62134 (C4 (Web Text))

Max Activating Token Index: 731

Text #12

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4579. Min Act: -0.0001

Data Index: 991464 (C4 (Web Text))

Max Activating Token Index: 457

Text #13

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4013. Min Act: -0.0001

Data Index: 1261936 (C4 (Web Text))

Max Activating Token Index: 248

Text #14

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4541. Min Act: -0.0002

Data Index: 364720 (C4 (Web Text))

Max Activating Token Index: 55

Text #15

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4354. Min Act: -0.0002

Data Index: 422403 (C4 (Web Text))

Max Activating Token Index: 39

Text #16

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4082. Min Act: -0.0002

Data Index: 877446 (C4 (Web Text))

Max Activating Token Index: 351

Text #17

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4494. Min Act: -0.0002

Data Index: 699576 (C4 (Web Text))

Max Activating Token Index: 538

Text #18

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4179. Min Act: -0.0002

Data Index: 796467 (C4 (Web Text))

Max Activating Token Index: 32

Text #19

Max Range: 0.6447. Min Range: -0.6447

Max Act: 0.4048. Min Act: -0.0002

Data Index: 581078 (C4 (Web Text))

Max Activating Token Index: 368