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

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

Neuron 4052 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: 0.5221. Min Range: -0.5221

Max Act: 0.5069. Min Act: -0.0001

Data Index: 931381 (C4 (Web Text))

Max Activating Token Index: 547

Text #1

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.5138. Min Act: -0.0001

Data Index: 748077 (C4 (Web Text))

Max Activating Token Index: 623

Text #2

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.5004. Min Act: -0.0001

Data Index: 1220139 (C4 (Web Text))

Max Activating Token Index: 501

Text #3

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.5221. Min Act: -0.0001

Data Index: 265409 (C4 (Web Text))

Max Activating Token Index: 124

Text #4

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4628. Min Act: -0.0001

Data Index: 308976 (C4 (Web Text))

Max Activating Token Index: 672

Text #5

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4935. Min Act: -0.0001

Data Index: 230072 (C4 (Web Text))

Max Activating Token Index: 772

Text #6

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4597. Min Act: -0.0001

Data Index: 940865 (C4 (Web Text))

Max Activating Token Index: 941

Text #7

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.5089. Min Act: -0.0001

Data Index: 1495011 (Python Code)

Max Activating Token Index: 793

Text #8

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4440. Min Act: -0.0001

Data Index: 663502 (C4 (Web Text))

Max Activating Token Index: 516

Text #9

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4716. Min Act: -0.0001

Data Index: 166586 (C4 (Web Text))

Max Activating Token Index: 199

Text #10

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4737. Min Act: -0.0001

Data Index: 654166 (C4 (Web Text))

Max Activating Token Index: 538

Text #11

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4404. Min Act: -0.0001

Data Index: 434677 (C4 (Web Text))

Max Activating Token Index: 867

Text #12

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4234. Min Act: -0.0001

Data Index: 241304 (C4 (Web Text))

Max Activating Token Index: 408

Text #13

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4677. Min Act: -0.0001

Data Index: 1513213 (Python Code)

Max Activating Token Index: 227

Text #14

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4689. Min Act: -0.0001

Data Index: 1297643 (C4 (Web Text))

Max Activating Token Index: 625

Text #15

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4844. Min Act: -0.0001

Data Index: 881475 (C4 (Web Text))

Max Activating Token Index: 711

Text #16

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4683. Min Act: -0.0001

Data Index: 1263998 (C4 (Web Text))

Max Activating Token Index: 600

Text #17

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4340. Min Act: -0.0001

Data Index: 1017249 (C4 (Web Text))

Max Activating Token Index: 415

Text #18

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4322. Min Act: -0.0001

Data Index: 483389 (C4 (Web Text))

Max Activating Token Index: 952

Text #19

Max Range: 0.5221. Min Range: -0.5221

Max Act: 0.4111. Min Act: -0.0001

Data Index: 853597 (C4 (Web Text))

Max Activating Token Index: 981