Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3107 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.6335. Min Act: -0.0000

Data Index: 1638742 (Python Code)

Max Activating Token Index: 357

Text #1

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.6450. Min Act: -0.0000

Data Index: 262538 (C4 (Web Text))

Max Activating Token Index: 627

Text #2

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.6325. Min Act: 0.0000

Data Index: 1622048 (Python Code)

Max Activating Token Index: 313

Text #3

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.5705. Min Act: -0.0000

Data Index: 1148607 (C4 (Web Text))

Max Activating Token Index: 750

Text #4

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.5322. Min Act: -0.0000

Data Index: 1490938 (Python Code)

Max Activating Token Index: 159

Text #5

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4580. Min Act: -0.0000

Data Index: 1540412 (Python Code)

Max Activating Token Index: 756

Text #6

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.5142. Min Act: -0.0000

Data Index: 1650669 (Python Code)

Max Activating Token Index: 284

Text #7

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.5131. Min Act: -0.0000

Data Index: 1478660 (Python Code)

Max Activating Token Index: 756

Text #8

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4839. Min Act: -0.0000

Data Index: 1545923 (Python Code)

Max Activating Token Index: 712

Text #9

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4564. Min Act: -0.0000

Data Index: 591736 (C4 (Web Text))

Max Activating Token Index: 320

Text #10

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4727. Min Act: -0.0000

Data Index: 809737 (C4 (Web Text))

Max Activating Token Index: 189

Text #11

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4611. Min Act: -0.0000

Data Index: 1478841 (Python Code)

Max Activating Token Index: 952

Text #12

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4674. Min Act: -0.0000

Data Index: 1630023 (Python Code)

Max Activating Token Index: 417

Text #13

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4462. Min Act: -0.0000

Data Index: 1640372 (Python Code)

Max Activating Token Index: 452

Text #14

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4508. Min Act: -0.0000

Data Index: 1441839 (Python Code)

Max Activating Token Index: 546

Text #15

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4136. Min Act: 0.0000

Data Index: 1474928 (Python Code)

Max Activating Token Index: 183

Text #16

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4310. Min Act: -0.0000

Data Index: 774960 (C4 (Web Text))

Max Activating Token Index: 925

Text #17

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4351. Min Act: -0.0000

Data Index: 487108 (C4 (Web Text))

Max Activating Token Index: 215

Text #18

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4485. Min Act: -0.0000

Data Index: 527780 (C4 (Web Text))

Max Activating Token Index: 744

Text #19

Max Range: 0.6450. Min Range: -0.6450

Max Act: 0.4448. Min Act: -0.0000

Data Index: 1426879 (Python Code)

Max Activating Token Index: 819