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

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

Neuron 4787 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.6549. Min Range: -4.6549

Max Act: 4.6549. Min Act: -0.0000

Data Index: 1451368 (Python Code)

Max Activating Token Index: 168

Text #1

Max Range: 4.6549. Min Range: -4.6549

Max Act: 4.0892. Min Act: -0.0000

Data Index: 1371556 (Python Code)

Max Activating Token Index: 42

Text #2

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.8682. Min Act: -0.0000

Data Index: 998532 (C4 (Web Text))

Max Activating Token Index: 30

Text #3

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.7233. Min Act: -0.0000

Data Index: 1044816 (C4 (Web Text))

Max Activating Token Index: 414

Text #4

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.7962. Min Act: -0.0000

Data Index: 307804 (C4 (Web Text))

Max Activating Token Index: 286

Text #5

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.6407. Min Act: -0.0000

Data Index: 1557091 (Python Code)

Max Activating Token Index: 913

Text #6

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.7985. Min Act: -0.0000

Data Index: 296827 (C4 (Web Text))

Max Activating Token Index: 83

Text #7

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.4447. Min Act: -0.0000

Data Index: 1496015 (Python Code)

Max Activating Token Index: 926

Text #8

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.5504. Min Act: -0.0000

Data Index: 944623 (C4 (Web Text))

Max Activating Token Index: 708

Text #9

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.3849. Min Act: -0.0000

Data Index: 725988 (C4 (Web Text))

Max Activating Token Index: 189

Text #10

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.4952. Min Act: -0.0000

Data Index: 1077938 (C4 (Web Text))

Max Activating Token Index: 505

Text #11

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.3321. Min Act: -0.0000

Data Index: 1507773 (Python Code)

Max Activating Token Index: 703

Text #12

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.1705. Min Act: -0.0000

Data Index: 302020 (C4 (Web Text))

Max Activating Token Index: 187

Text #13

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.8202. Min Act: -0.0000

Data Index: 647006 (C4 (Web Text))

Max Activating Token Index: 242

Text #14

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.3168. Min Act: -0.0000

Data Index: 1388044 (Python Code)

Max Activating Token Index: 541

Text #15

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.6508. Min Act: -0.0000

Data Index: 715077 (C4 (Web Text))

Max Activating Token Index: 194

Text #16

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.4614. Min Act: -0.0000

Data Index: 1368317 (Python Code)

Max Activating Token Index: 830

Text #17

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.4036. Min Act: -0.0000

Data Index: 790509 (C4 (Web Text))

Max Activating Token Index: 464

Text #18

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.3072. Min Act: -0.0000

Data Index: 1387732 (Python Code)

Max Activating Token Index: 942

Text #19

Max Range: 4.6549. Min Range: -4.6549

Max Act: 3.3013. Min Act: -0.0000

Data Index: 1387697 (Python Code)

Max Activating Token Index: 81