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

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

Neuron 226 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1888. Min Range: -1.1888

Max Act: 1.1862. Min Act: -0.0000

Data Index: 951738 (C4 (Web Text))

Max Activating Token Index: 1014

Text #1

Max Range: 1.1888. Min Range: -1.1888

Max Act: 1.1520. Min Act: -0.0000

Data Index: 59835 (C4 (Web Text))

Max Activating Token Index: 676

Text #2

Max Range: 1.1888. Min Range: -1.1888

Max Act: 1.1888. Min Act: -0.0000

Data Index: 1055367 (C4 (Web Text))

Max Activating Token Index: 895

Text #3

Max Range: 1.1888. Min Range: -1.1888

Max Act: 1.1700. Min Act: 0.0001

Data Index: 744714 (C4 (Web Text))

Max Activating Token Index: 40

Text #4

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.9975. Min Act: -0.0000

Data Index: 740820 (C4 (Web Text))

Max Activating Token Index: 798

Text #5

Max Range: 1.1888. Min Range: -1.1888

Max Act: 1.1382. Min Act: -0.0001

Data Index: 1386363 (Python Code)

Max Activating Token Index: 796

Text #6

Max Range: 1.1888. Min Range: -1.1888

Max Act: 1.0538. Min Act: 0.0000

Data Index: 132660 (C4 (Web Text))

Max Activating Token Index: 253

Text #7

Max Range: 1.1888. Min Range: -1.1888

Max Act: 1.0803. Min Act: 0.0000

Data Index: 1398558 (Python Code)

Max Activating Token Index: 525

Text #8

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.9994. Min Act: -0.0000

Data Index: 1275254 (C4 (Web Text))

Max Activating Token Index: 728

Text #9

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.9366. Min Act: 0.0000

Data Index: 903851 (C4 (Web Text))

Max Activating Token Index: 13

Text #10

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.9840. Min Act: -0.0000

Data Index: 1249190 (C4 (Web Text))

Max Activating Token Index: 912

Text #11

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.9306. Min Act: -0.0001

Data Index: 821805 (C4 (Web Text))

Max Activating Token Index: 475

Text #12

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.9257. Min Act: -0.0001

Data Index: 1495933 (Python Code)

Max Activating Token Index: 989

Text #13

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.9182. Min Act: -0.0000

Data Index: 197147 (C4 (Web Text))

Max Activating Token Index: 185

Text #14

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.6220. Min Act: -0.0000

Data Index: 1012285 (C4 (Web Text))

Max Activating Token Index: 922

Text #15

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.8557. Min Act: -0.0000

Data Index: 790912 (C4 (Web Text))

Max Activating Token Index: 305

Text #16

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.8495. Min Act: -0.0000

Data Index: 1239495 (C4 (Web Text))

Max Activating Token Index: 194

Text #17

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.7845. Min Act: -0.0000

Data Index: 351908 (C4 (Web Text))

Max Activating Token Index: 879

Text #18

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.7948. Min Act: 0.0000

Data Index: 233425 (C4 (Web Text))

Max Activating Token Index: 178

Text #19

Max Range: 1.1888. Min Range: -1.1888

Max Act: 0.8725. Min Act: -0.0000

Data Index: 1397135 (Python Code)

Max Activating Token Index: 840