Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 20 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2755. Min Act: -0.0000

Data Index: 659838 (C4 (Web Text))

Max Activating Token Index: 200

Text #1

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2656. Min Act: -0.0001

Data Index: 10404 (C4 (Web Text))

Max Activating Token Index: 57

Text #2

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2688. Min Act: -0.0000

Data Index: 1059853 (C4 (Web Text))

Max Activating Token Index: 916

Text #3

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2523. Min Act: -0.0000

Data Index: 786904 (C4 (Web Text))

Max Activating Token Index: 1012

Text #4

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2585. Min Act: -0.0000

Data Index: 1082571 (C4 (Web Text))

Max Activating Token Index: 815

Text #5

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2474. Min Act: -0.0001

Data Index: 1031902 (C4 (Web Text))

Max Activating Token Index: 104

Text #6

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2434. Min Act: -0.0000

Data Index: 114085 (C4 (Web Text))

Max Activating Token Index: 447

Text #7

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2492. Min Act: -0.0000

Data Index: 913501 (C4 (Web Text))

Max Activating Token Index: 932

Text #8

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2375. Min Act: -0.0000

Data Index: 786426 (C4 (Web Text))

Max Activating Token Index: 474

Text #9

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2317. Min Act: -0.0000

Data Index: 1139620 (C4 (Web Text))

Max Activating Token Index: 929

Text #10

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2125. Min Act: -0.0001

Data Index: 605318 (C4 (Web Text))

Max Activating Token Index: 609

Text #11

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2340. Min Act: -0.0000

Data Index: 1211784 (C4 (Web Text))

Max Activating Token Index: 551

Text #12

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2241. Min Act: -0.0001

Data Index: 732011 (C4 (Web Text))

Max Activating Token Index: 1017

Text #13

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2233. Min Act: -0.0000

Data Index: 280225 (C4 (Web Text))

Max Activating Token Index: 242

Text #14

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2293. Min Act: -0.0001

Data Index: 1412517 (Python Code)

Max Activating Token Index: 159

Text #15

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2216. Min Act: -0.0001

Data Index: 802674 (C4 (Web Text))

Max Activating Token Index: 787

Text #16

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2206. Min Act: -0.0000

Data Index: 238635 (C4 (Web Text))

Max Activating Token Index: 233

Text #17

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2350. Min Act: -0.0001

Data Index: 799844 (C4 (Web Text))

Max Activating Token Index: 575

Text #18

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2173. Min Act: -0.0001

Data Index: 1141912 (C4 (Web Text))

Max Activating Token Index: 411

Text #19

Max Range: 0.2755. Min Range: -0.2755

Max Act: 0.2107. Min Act: -0.0001

Data Index: 626638 (C4 (Web Text))

Max Activating Token Index: 632