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

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

Neuron 2926 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.7515. Min Act: -0.0000

Data Index: 581222 (C4 (Web Text))

Max Activating Token Index: 863

Text #1

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4906. Min Act: -0.0000

Data Index: 1535887 (Python Code)

Max Activating Token Index: 55

Text #2

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.5030. Min Act: -0.0000

Data Index: 62934 (C4 (Web Text))

Max Activating Token Index: 110

Text #3

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.5066. Min Act: -0.0000

Data Index: 670965 (C4 (Web Text))

Max Activating Token Index: 85

Text #4

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.5206. Min Act: -0.0000

Data Index: 572774 (C4 (Web Text))

Max Activating Token Index: 524

Text #5

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.5168. Min Act: -0.0000

Data Index: 464258 (C4 (Web Text))

Max Activating Token Index: 294

Text #6

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4652. Min Act: -0.0000

Data Index: 664090 (C4 (Web Text))

Max Activating Token Index: 388

Text #7

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4552. Min Act: -0.0000

Data Index: 449127 (C4 (Web Text))

Max Activating Token Index: 498

Text #8

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4721. Min Act: -0.0000

Data Index: 1055860 (C4 (Web Text))

Max Activating Token Index: 291

Text #9

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4961. Min Act: -0.0000

Data Index: 173940 (C4 (Web Text))

Max Activating Token Index: 901

Text #10

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4757. Min Act: -0.0000

Data Index: 519821 (C4 (Web Text))

Max Activating Token Index: 369

Text #11

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4428. Min Act: -0.0000

Data Index: 803821 (C4 (Web Text))

Max Activating Token Index: 50

Text #12

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4391. Min Act: -0.0000

Data Index: 264425 (C4 (Web Text))

Max Activating Token Index: 581

Text #13

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4687. Min Act: -0.0000

Data Index: 648799 (C4 (Web Text))

Max Activating Token Index: 45

Text #14

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4622. Min Act: -0.0000

Data Index: 1392237 (Python Code)

Max Activating Token Index: 977

Text #15

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4576. Min Act: -0.0000

Data Index: 1316839 (C4 (Web Text))

Max Activating Token Index: 1006

Text #16

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4191. Min Act: -0.0000

Data Index: 87892 (C4 (Web Text))

Max Activating Token Index: 896

Text #17

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4096. Min Act: -0.0000

Data Index: 1361089 (Python Code)

Max Activating Token Index: 84

Text #18

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.3900. Min Act: -0.0000

Data Index: 1630316 (Python Code)

Max Activating Token Index: 113

Text #19

Max Range: 0.7515. Min Range: -0.7515

Max Act: 0.4198. Min Act: -0.0000

Data Index: 642865 (C4 (Web Text))

Max Activating Token Index: 107