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

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

Neuron 1540 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.6192. Min Act: -0.0001

Data Index: 347598 (C4 (Web Text))

Max Activating Token Index: 964

Text #1

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.5637. Min Act: -0.0001

Data Index: 221521 (C4 (Web Text))

Max Activating Token Index: 939

Text #2

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.6046. Min Act: -0.0001

Data Index: 585082 (C4 (Web Text))

Max Activating Token Index: 889

Text #3

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.5840. Min Act: -0.0001

Data Index: 1330495 (C4 (Web Text))

Max Activating Token Index: 203

Text #4

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.5940. Min Act: -0.0001

Data Index: 85098 (C4 (Web Text))

Max Activating Token Index: 1002

Text #5

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.4762. Min Act: -0.0001

Data Index: 673963 (C4 (Web Text))

Max Activating Token Index: 864

Text #6

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.4326. Min Act: -0.0001

Data Index: 816414 (C4 (Web Text))

Max Activating Token Index: 12

Text #7

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.4237. Min Act: -0.0000

Data Index: 1104886 (C4 (Web Text))

Max Activating Token Index: 212

Text #8

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.4117. Min Act: -0.0000

Data Index: 643014 (C4 (Web Text))

Max Activating Token Index: 73

Text #9

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.4202. Min Act: -0.0001

Data Index: 213231 (C4 (Web Text))

Max Activating Token Index: 314

Text #10

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.3982. Min Act: -0.0001

Data Index: 1251593 (C4 (Web Text))

Max Activating Token Index: 942

Text #11

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.4021. Min Act: -0.0001

Data Index: 164704 (C4 (Web Text))

Max Activating Token Index: 738

Text #12

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.3640. Min Act: -0.0000

Data Index: 380236 (C4 (Web Text))

Max Activating Token Index: 447

Text #13

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.3599. Min Act: -0.0001

Data Index: 1061496 (C4 (Web Text))

Max Activating Token Index: 242

Text #14

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.3729. Min Act: -0.0000

Data Index: 1143338 (C4 (Web Text))

Max Activating Token Index: 290

Text #15

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.3489. Min Act: -0.0001

Data Index: 245851 (C4 (Web Text))

Max Activating Token Index: 365

Text #16

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.3361. Min Act: -0.0000

Data Index: 1422434 (Python Code)

Max Activating Token Index: 116

Text #17

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.3493. Min Act: -0.0001

Data Index: 1320083 (C4 (Web Text))

Max Activating Token Index: 533

Text #18

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.3308. Min Act: -0.0001

Data Index: 1182849 (C4 (Web Text))

Max Activating Token Index: 787

Text #19

Max Range: 0.6192. Min Range: -0.6192

Max Act: 0.3327. Min Act: -0.0001

Data Index: 575379 (C4 (Web Text))

Max Activating Token Index: 63