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

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

Neuron 4438 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7292. Min Range: -2.7292

Max Act: 2.7292. Min Act: -0.0000

Data Index: 271476 (C4 (Web Text))

Max Activating Token Index: 11

Text #1

Max Range: 2.7292. Min Range: -2.7292

Max Act: 2.6724. Min Act: -0.0000

Data Index: 52285 (C4 (Web Text))

Max Activating Token Index: 2

Text #2

Max Range: 2.7292. Min Range: -2.7292

Max Act: 2.5404. Min Act: -0.0000

Data Index: 493163 (C4 (Web Text))

Max Activating Token Index: 784

Text #3

Max Range: 2.7292. Min Range: -2.7292

Max Act: 2.4624. Min Act: -0.0000

Data Index: 1277524 (C4 (Web Text))

Max Activating Token Index: 623

Text #4

Max Range: 2.7292. Min Range: -2.7292

Max Act: 2.1834. Min Act: -0.0000

Data Index: 356641 (C4 (Web Text))

Max Activating Token Index: 792

Text #5

Max Range: 2.7292. Min Range: -2.7292

Max Act: 2.3231. Min Act: -0.0000

Data Index: 1402277 (Python Code)

Max Activating Token Index: 807

Text #6

Max Range: 2.7292. Min Range: -2.7292

Max Act: 2.2293. Min Act: -0.0000

Data Index: 599217 (C4 (Web Text))

Max Activating Token Index: 847

Text #7

Max Range: 2.7292. Min Range: -2.7292

Max Act: 2.0228. Min Act: -0.0000

Data Index: 816349 (C4 (Web Text))

Max Activating Token Index: 65

Text #8

Max Range: 2.7292. Min Range: -2.7292

Max Act: 2.1751. Min Act: -0.0000

Data Index: 1228405 (C4 (Web Text))

Max Activating Token Index: 763

Text #9

Max Range: 2.7292. Min Range: -2.7292

Max Act: 2.0293. Min Act: -0.0000

Data Index: 784865 (C4 (Web Text))

Max Activating Token Index: 268

Text #10

Max Range: 2.7292. Min Range: -2.7292

Max Act: 2.0089. Min Act: -0.0000

Data Index: 1603145 (Python Code)

Max Activating Token Index: 611

Text #11

Max Range: 2.7292. Min Range: -2.7292

Max Act: 1.9998. Min Act: -0.0000

Data Index: 1266187 (C4 (Web Text))

Max Activating Token Index: 1007

Text #12

Max Range: 2.7292. Min Range: -2.7292

Max Act: 1.9556. Min Act: -0.0000

Data Index: 165143 (C4 (Web Text))

Max Activating Token Index: 909

Text #13

Max Range: 2.7292. Min Range: -2.7292

Max Act: 1.8150. Min Act: -0.0000

Data Index: 1166541 (C4 (Web Text))

Max Activating Token Index: 479

Text #14

Max Range: 2.7292. Min Range: -2.7292

Max Act: 1.9054. Min Act: -0.0000

Data Index: 883011 (C4 (Web Text))

Max Activating Token Index: 447

Text #15

Max Range: 2.7292. Min Range: -2.7292

Max Act: 1.9280. Min Act: -0.0000

Data Index: 495095 (C4 (Web Text))

Max Activating Token Index: 689

Text #16

Max Range: 2.7292. Min Range: -2.7292

Max Act: 1.8846. Min Act: -0.0000

Data Index: 437767 (C4 (Web Text))

Max Activating Token Index: 251

Text #17

Max Range: 2.7292. Min Range: -2.7292

Max Act: 1.8127. Min Act: -0.0000

Data Index: 1286856 (C4 (Web Text))

Max Activating Token Index: 189

Text #18

Max Range: 2.7292. Min Range: -2.7292

Max Act: 1.4935. Min Act: -0.0000

Data Index: 264892 (C4 (Web Text))

Max Activating Token Index: 578

Text #19

Max Range: 2.7292. Min Range: -2.7292

Max Act: 1.7274. Min Act: -0.0000

Data Index: 589380 (C4 (Web Text))

Max Activating Token Index: 230