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

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

Neuron 5204 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: 1.4940. Min Range: -1.4940

Max Act: 1.4940. Min Act: -0.0000

Data Index: 973518 (C4 (Web Text))

Max Activating Token Index: 61

Text #1

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.4882. Min Act: -0.0000

Data Index: 598416 (C4 (Web Text))

Max Activating Token Index: 63

Text #2

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.4409. Min Act: -0.0000

Data Index: 320410 (C4 (Web Text))

Max Activating Token Index: 914

Text #3

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.3788. Min Act: -0.0000

Data Index: 919897 (C4 (Web Text))

Max Activating Token Index: 794

Text #4

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.3682. Min Act: -0.0000

Data Index: 639079 (C4 (Web Text))

Max Activating Token Index: 959

Text #5

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.4055. Min Act: -0.0000

Data Index: 637227 (C4 (Web Text))

Max Activating Token Index: 964

Text #6

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.4459. Min Act: -0.0000

Data Index: 1360563 (Python Code)

Max Activating Token Index: 1015

Text #7

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.3424. Min Act: -0.0000

Data Index: 86599 (C4 (Web Text))

Max Activating Token Index: 799

Text #8

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.2661. Min Act: -0.0000

Data Index: 22518 (C4 (Web Text))

Max Activating Token Index: 679

Text #9

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.2910. Min Act: -0.0000

Data Index: 530352 (C4 (Web Text))

Max Activating Token Index: 912

Text #10

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.2868. Min Act: -0.0000

Data Index: 1034818 (C4 (Web Text))

Max Activating Token Index: 948

Text #11

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.3409. Min Act: -0.0000

Data Index: 1457826 (Python Code)

Max Activating Token Index: 964

Text #12

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.3297. Min Act: -0.0000

Data Index: 981491 (C4 (Web Text))

Max Activating Token Index: 790

Text #13

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.3440. Min Act: -0.0000

Data Index: 1065371 (C4 (Web Text))

Max Activating Token Index: 642

Text #14

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.2918. Min Act: -0.0000

Data Index: 1214965 (C4 (Web Text))

Max Activating Token Index: 912

Text #15

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.2857. Min Act: -0.0000

Data Index: 1523742 (Python Code)

Max Activating Token Index: 749

Text #16

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.3202. Min Act: -0.0000

Data Index: 802352 (C4 (Web Text))

Max Activating Token Index: 957

Text #17

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.2351. Min Act: -0.0000

Data Index: 1227922 (C4 (Web Text))

Max Activating Token Index: 536

Text #18

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.2761. Min Act: -0.0000

Data Index: 261113 (C4 (Web Text))

Max Activating Token Index: 365

Text #19

Max Range: 1.4940. Min Range: -1.4940

Max Act: 1.2387. Min Act: -0.0000

Data Index: 51324 (C4 (Web Text))

Max Activating Token Index: 542