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

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

Neuron 2387 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.3038. Min Range: -2.3038

Max Act: 2.3038. Min Act: -0.0001

Data Index: 1410558 (Python Code)

Max Activating Token Index: 22

Text #1

Max Range: 2.3038. Min Range: -2.3038

Max Act: 2.1378. Min Act: -0.0001

Data Index: 1584837 (Python Code)

Max Activating Token Index: 31

Text #2

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.7993. Min Act: -0.0001

Data Index: 315869 (C4 (Web Text))

Max Activating Token Index: 465

Text #3

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.7426. Min Act: -0.0001

Data Index: 1360609 (Python Code)

Max Activating Token Index: 618

Text #4

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.8205. Min Act: -0.0000

Data Index: 1387815 (Python Code)

Max Activating Token Index: 171

Text #5

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.7469. Min Act: -0.0001

Data Index: 1275425 (C4 (Web Text))

Max Activating Token Index: 810

Text #6

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.6215. Min Act: -0.0001

Data Index: 1188935 (C4 (Web Text))

Max Activating Token Index: 352

Text #7

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.8337. Min Act: -0.0001

Data Index: 1451343 (Python Code)

Max Activating Token Index: 708

Text #8

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.6163. Min Act: -0.0001

Data Index: 149421 (C4 (Web Text))

Max Activating Token Index: 274

Text #9

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.8217. Min Act: -0.0001

Data Index: 855041 (C4 (Web Text))

Max Activating Token Index: 103

Text #10

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.8149. Min Act: -0.0001

Data Index: 1056688 (C4 (Web Text))

Max Activating Token Index: 443

Text #11

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.7090. Min Act: -0.0001

Data Index: 727747 (C4 (Web Text))

Max Activating Token Index: 100

Text #12

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.7158. Min Act: -0.0001

Data Index: 726465 (C4 (Web Text))

Max Activating Token Index: 961

Text #13

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.9613. Min Act: -0.0001

Data Index: 1383178 (Python Code)

Max Activating Token Index: 100

Text #14

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.8379. Min Act: -0.0001

Data Index: 353177 (C4 (Web Text))

Max Activating Token Index: 508

Text #15

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.5846. Min Act: -0.0001

Data Index: 231864 (C4 (Web Text))

Max Activating Token Index: 720

Text #16

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.6842. Min Act: -0.0000

Data Index: 1463903 (Python Code)

Max Activating Token Index: 7

Text #17

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.6842. Min Act: -0.0000

Data Index: 1636796 (Python Code)

Max Activating Token Index: 7

Text #18

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.8308. Min Act: -0.0000

Data Index: 1489028 (Python Code)

Max Activating Token Index: 160

Text #19

Max Range: 2.3038. Min Range: -2.3038

Max Act: 1.7677. Min Act: -0.0001

Data Index: 918608 (C4 (Web Text))

Max Activating Token Index: 233