Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 1770 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.1396. Min Range: -1.1396

Max Act: 1.0531. Min Act: -0.0001

Data Index: 701324 (C4 (Web Text))

Max Activating Token Index: 355

Text #1

Max Range: 1.1396. Min Range: -1.1396

Max Act: 1.1396. Min Act: -0.0001

Data Index: 190217 (C4 (Web Text))

Max Activating Token Index: 311

Text #2

Max Range: 1.1396. Min Range: -1.1396

Max Act: 1.0246. Min Act: -0.0001

Data Index: 97296 (C4 (Web Text))

Max Activating Token Index: 137

Text #3

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.8084. Min Act: -0.0001

Data Index: 968354 (C4 (Web Text))

Max Activating Token Index: 101

Text #4

Max Range: 1.1396. Min Range: -1.1396

Max Act: 1.0464. Min Act: -0.0001

Data Index: 1316472 (C4 (Web Text))

Max Activating Token Index: 460

Text #5

Max Range: 1.1396. Min Range: -1.1396

Max Act: 1.0073. Min Act: -0.0001

Data Index: 454083 (C4 (Web Text))

Max Activating Token Index: 263

Text #6

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.9264. Min Act: -0.0001

Data Index: 261167 (C4 (Web Text))

Max Activating Token Index: 734

Text #7

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.9507. Min Act: -0.0001

Data Index: 822203 (C4 (Web Text))

Max Activating Token Index: 4

Text #8

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.9659. Min Act: -0.0001

Data Index: 921764 (C4 (Web Text))

Max Activating Token Index: 976

Text #9

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.9420. Min Act: -0.0001

Data Index: 1619935 (Python Code)

Max Activating Token Index: 21

Text #10

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.9402. Min Act: -0.0001

Data Index: 1602692 (Python Code)

Max Activating Token Index: 108

Text #11

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.9009. Min Act: -0.0001

Data Index: 1452317 (Python Code)

Max Activating Token Index: 806

Text #12

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.8996. Min Act: -0.0001

Data Index: 1366402 (Python Code)

Max Activating Token Index: 808

Text #13

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.8356. Min Act: -0.0001

Data Index: 604169 (C4 (Web Text))

Max Activating Token Index: 143

Text #14

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.9921. Min Act: -0.0001

Data Index: 747716 (C4 (Web Text))

Max Activating Token Index: 217

Text #15

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.9515. Min Act: -0.0001

Data Index: 65314 (C4 (Web Text))

Max Activating Token Index: 749

Text #16

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.7605. Min Act: -0.0001

Data Index: 1266359 (C4 (Web Text))

Max Activating Token Index: 526

Text #17

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.9217. Min Act: -0.0001

Data Index: 838352 (C4 (Web Text))

Max Activating Token Index: 357

Text #18

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.9229. Min Act: -0.0001

Data Index: 1383612 (Python Code)

Max Activating Token Index: 159

Text #19

Max Range: 1.1396. Min Range: -1.1396

Max Act: 0.9135. Min Act: -0.0001

Data Index: 47843 (C4 (Web Text))

Max Activating Token Index: 186