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

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

Neuron 1501 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.1141. Min Act: -0.0001

Data Index: 1093572 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0610. Min Act: -0.0001

Data Index: 59258 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0627. Min Act: -0.0001

Data Index: 1298876 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0501. Min Act: -0.0001

Data Index: 261336 (C4 (Web Text))

Max Activating Token Index: 863

Text #4

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0501. Min Act: -0.0001

Data Index: 387183 (C4 (Web Text))

Max Activating Token Index: 964

Text #5

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0461. Min Act: -0.0001

Data Index: 1649072 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0427. Min Act: -0.0001

Data Index: 44532 (C4 (Web Text))

Max Activating Token Index: 875

Text #7

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0430. Min Act: -0.0001

Data Index: 335531 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0385. Min Act: -0.0001

Data Index: 1595780 (Python Code)

Max Activating Token Index: 1023

Text #9

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0347. Min Act: -0.0001

Data Index: 85954 (C4 (Web Text))

Max Activating Token Index: 575

Text #10

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0355. Min Act: -0.0001

Data Index: 1233715 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0340. Min Act: -0.0001

Data Index: 1431532 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0331. Min Act: -0.0001

Data Index: 564083 (C4 (Web Text))

Max Activating Token Index: 668

Text #13

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0368. Min Act: -0.0001

Data Index: 773102 (C4 (Web Text))

Max Activating Token Index: 727

Text #14

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0316. Min Act: -0.0001

Data Index: 1388406 (Python Code)

Max Activating Token Index: 960

Text #15

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0336. Min Act: -0.0001

Data Index: 1499762 (Python Code)

Max Activating Token Index: 939

Text #16

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0318. Min Act: -0.0001

Data Index: 1141712 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0288. Min Act: -0.0001

Data Index: 1351930 (C4 (Web Text))

Max Activating Token Index: 978

Text #18

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0304. Min Act: -0.0001

Data Index: 1373935 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 0.1141. Min Range: -0.1141

Max Act: 0.0323. Min Act: -0.0001

Data Index: 757840 (C4 (Web Text))

Max Activating Token Index: 678