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

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

Neuron 2097 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.5461. Min Act: -0.0000

Data Index: 186441 (C4 (Web Text))

Max Activating Token Index: 26

Text #1

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.5515. Min Act: -0.0000

Data Index: 783833 (C4 (Web Text))

Max Activating Token Index: 532

Text #2

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.4252. Min Act: -0.0000

Data Index: 43075 (C4 (Web Text))

Max Activating Token Index: 909

Text #3

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.4440. Min Act: -0.0000

Data Index: 643167 (C4 (Web Text))

Max Activating Token Index: 378

Text #4

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.4625. Min Act: -0.0000

Data Index: 615199 (C4 (Web Text))

Max Activating Token Index: 775

Text #5

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.4644. Min Act: -0.0000

Data Index: 17261 (C4 (Web Text))

Max Activating Token Index: 238

Text #6

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.4298. Min Act: -0.0000

Data Index: 147936 (C4 (Web Text))

Max Activating Token Index: 997

Text #7

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.3940. Min Act: -0.0000

Data Index: 613182 (C4 (Web Text))

Max Activating Token Index: 334

Text #8

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.4077. Min Act: -0.0000

Data Index: 693169 (C4 (Web Text))

Max Activating Token Index: 774

Text #9

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.4089. Min Act: -0.0000

Data Index: 664489 (C4 (Web Text))

Max Activating Token Index: 430

Text #10

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.4188. Min Act: -0.0000

Data Index: 185942 (C4 (Web Text))

Max Activating Token Index: 59

Text #11

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.3157. Min Act: -0.0000

Data Index: 1462420 (Python Code)

Max Activating Token Index: 115

Text #12

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.4303. Min Act: -0.0000

Data Index: 30233 (C4 (Web Text))

Max Activating Token Index: 646

Text #13

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.3713. Min Act: -0.0000

Data Index: 800126 (C4 (Web Text))

Max Activating Token Index: 812

Text #14

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.3877. Min Act: -0.0000

Data Index: 1122218 (C4 (Web Text))

Max Activating Token Index: 792

Text #15

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.4020. Min Act: -0.0000

Data Index: 932942 (C4 (Web Text))

Max Activating Token Index: 108

Text #16

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.3846. Min Act: -0.0000

Data Index: 1556387 (Python Code)

Max Activating Token Index: 23

Text #17

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.3751. Min Act: -0.0000

Data Index: 664090 (C4 (Web Text))

Max Activating Token Index: 351

Text #18

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.3775. Min Act: -0.0000

Data Index: 498920 (C4 (Web Text))

Max Activating Token Index: 536

Text #19

Max Range: 0.5515. Min Range: -0.5515

Max Act: 0.3739. Min Act: -0.0000

Data Index: 1046455 (C4 (Web Text))

Max Activating Token Index: 661