Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2478 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6935. Min Act: -0.0003

Data Index: 116273 (C4 (Web Text))

Max Activating Token Index: 612

Text #1

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.7161. Min Act: -0.0003

Data Index: 340842 (C4 (Web Text))

Max Activating Token Index: 238

Text #2

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.7026. Min Act: -0.0003

Data Index: 476565 (C4 (Web Text))

Max Activating Token Index: 132

Text #3

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6689. Min Act: -0.0003

Data Index: 752695 (C4 (Web Text))

Max Activating Token Index: 178

Text #4

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6867. Min Act: -0.0003

Data Index: 937162 (C4 (Web Text))

Max Activating Token Index: 777

Text #5

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.7026. Min Act: -0.0003

Data Index: 1113029 (C4 (Web Text))

Max Activating Token Index: 198

Text #6

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6730. Min Act: -0.0003

Data Index: 1053468 (C4 (Web Text))

Max Activating Token Index: 615

Text #7

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6666. Min Act: -0.0003

Data Index: 902320 (C4 (Web Text))

Max Activating Token Index: 576

Text #8

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6389. Min Act: -0.0003

Data Index: 268225 (C4 (Web Text))

Max Activating Token Index: 147

Text #9

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6429. Min Act: -0.0003

Data Index: 475850 (C4 (Web Text))

Max Activating Token Index: 518

Text #10

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6469. Min Act: -0.0003

Data Index: 1356893 (C4 (Web Text))

Max Activating Token Index: 735

Text #11

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6281. Min Act: -0.0003

Data Index: 777064 (C4 (Web Text))

Max Activating Token Index: 754

Text #12

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6288. Min Act: -0.0003

Data Index: 1259432 (C4 (Web Text))

Max Activating Token Index: 566

Text #13

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6239. Min Act: -0.0003

Data Index: 969776 (C4 (Web Text))

Max Activating Token Index: 307

Text #14

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.5795. Min Act: -0.0003

Data Index: 170471 (C4 (Web Text))

Max Activating Token Index: 592

Text #15

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6022. Min Act: -0.0003

Data Index: 216324 (C4 (Web Text))

Max Activating Token Index: 557

Text #16

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6122. Min Act: -0.0003

Data Index: 1127991 (C4 (Web Text))

Max Activating Token Index: 507

Text #17

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.5905. Min Act: -0.0003

Data Index: 624104 (C4 (Web Text))

Max Activating Token Index: 227

Text #18

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.5789. Min Act: -0.0003

Data Index: 531530 (C4 (Web Text))

Max Activating Token Index: 875

Text #19

Max Range: 0.7161. Min Range: -0.7161

Max Act: 0.6289. Min Act: -0.0003

Data Index: 198872 (C4 (Web Text))

Max Activating Token Index: 474