Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5778 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 4.7339. Min Range: -4.7339

Max Act: 4.7339. Min Act: -0.0000

Data Index: 1621694 (Python Code)

Max Activating Token Index: 234

Text #1

Max Range: 4.7339. Min Range: -4.7339

Max Act: 4.4593. Min Act: -0.0000

Data Index: 1369005 (Python Code)

Max Activating Token Index: 568

Text #2

Max Range: 4.7339. Min Range: -4.7339

Max Act: 4.0631. Min Act: -0.0000

Data Index: 1572943 (Python Code)

Max Activating Token Index: 439

Text #3

Max Range: 4.7339. Min Range: -4.7339

Max Act: 4.1242. Min Act: -0.0000

Data Index: 205452 (C4 (Web Text))

Max Activating Token Index: 821

Text #4

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.7544. Min Act: -0.0000

Data Index: 1394997 (Python Code)

Max Activating Token Index: 930

Text #5

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.6857. Min Act: -0.0000

Data Index: 839805 (C4 (Web Text))

Max Activating Token Index: 821

Text #6

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.7289. Min Act: -0.0000

Data Index: 298796 (C4 (Web Text))

Max Activating Token Index: 378

Text #7

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.5822. Min Act: -0.0000

Data Index: 1172611 (C4 (Web Text))

Max Activating Token Index: 901

Text #8

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.6492. Min Act: -0.0000

Data Index: 131099 (C4 (Web Text))

Max Activating Token Index: 193

Text #9

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.6488. Min Act: -0.0000

Data Index: 1559781 (Python Code)

Max Activating Token Index: 976

Text #10

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.6151. Min Act: -0.0000

Data Index: 79375 (C4 (Web Text))

Max Activating Token Index: 803

Text #11

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.6035. Min Act: -0.0000

Data Index: 236274 (C4 (Web Text))

Max Activating Token Index: 710

Text #12

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.4633. Min Act: -0.0000

Data Index: 1563764 (Python Code)

Max Activating Token Index: 144

Text #13

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.7150. Min Act: -0.0000

Data Index: 1418866 (Python Code)

Max Activating Token Index: 166

Text #14

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.5654. Min Act: -0.0000

Data Index: 1169855 (C4 (Web Text))

Max Activating Token Index: 838

Text #15

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.5355. Min Act: -0.0000

Data Index: 1546940 (Python Code)

Max Activating Token Index: 99

Text #16

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.6959. Min Act: -0.0000

Data Index: 1058420 (C4 (Web Text))

Max Activating Token Index: 98

Text #17

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.6769. Min Act: -0.0000

Data Index: 1649624 (Python Code)

Max Activating Token Index: 772

Text #18

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.5202. Min Act: -0.0000

Data Index: 182536 (C4 (Web Text))

Max Activating Token Index: 181

Text #19

Max Range: 4.7339. Min Range: -4.7339

Max Act: 3.3938. Min Act: -0.0000

Data Index: 1307693 (C4 (Web Text))

Max Activating Token Index: 749