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

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

Neuron 5814 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4217. Min Range: -1.4217

Max Act: 1.4217. Min Act: -0.0000

Data Index: 472395 (C4 (Web Text))

Max Activating Token Index: 961

Text #1

Max Range: 1.4217. Min Range: -1.4217

Max Act: 1.1341. Min Act: -0.0000

Data Index: 419797 (C4 (Web Text))

Max Activating Token Index: 177

Text #2

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.9289. Min Act: -0.0000

Data Index: 772286 (C4 (Web Text))

Max Activating Token Index: 573

Text #3

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.8585. Min Act: -0.0000

Data Index: 519808 (C4 (Web Text))

Max Activating Token Index: 754

Text #4

Max Range: 1.4217. Min Range: -1.4217

Max Act: 1.0582. Min Act: -0.0000

Data Index: 510254 (C4 (Web Text))

Max Activating Token Index: 728

Text #5

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.9396. Min Act: -0.0000

Data Index: 775894 (C4 (Web Text))

Max Activating Token Index: 57

Text #6

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.9265. Min Act: -0.0000

Data Index: 1387296 (Python Code)

Max Activating Token Index: 8

Text #7

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.8630. Min Act: -0.0000

Data Index: 998835 (C4 (Web Text))

Max Activating Token Index: 42

Text #8

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.8977. Min Act: -0.0000

Data Index: 705382 (C4 (Web Text))

Max Activating Token Index: 1017

Text #9

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.8604. Min Act: -0.0000

Data Index: 394085 (C4 (Web Text))

Max Activating Token Index: 211

Text #10

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.8947. Min Act: -0.0000

Data Index: 353763 (C4 (Web Text))

Max Activating Token Index: 198

Text #11

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.8944. Min Act: -0.0000

Data Index: 588187 (C4 (Web Text))

Max Activating Token Index: 814

Text #12

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.8304. Min Act: -0.0000

Data Index: 833841 (C4 (Web Text))

Max Activating Token Index: 12

Text #13

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.8642. Min Act: -0.0000

Data Index: 200692 (C4 (Web Text))

Max Activating Token Index: 123

Text #14

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.8095. Min Act: -0.0000

Data Index: 959774 (C4 (Web Text))

Max Activating Token Index: 298

Text #15

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.8144. Min Act: -0.0000

Data Index: 1026054 (C4 (Web Text))

Max Activating Token Index: 191

Text #16

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.7373. Min Act: -0.0000

Data Index: 339372 (C4 (Web Text))

Max Activating Token Index: 162

Text #17

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.8451. Min Act: -0.0001

Data Index: 1548483 (Python Code)

Max Activating Token Index: 223

Text #18

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.7922. Min Act: -0.0000

Data Index: 1163935 (C4 (Web Text))

Max Activating Token Index: 681

Text #19

Max Range: 1.4217. Min Range: -1.4217

Max Act: 0.8397. Min Act: -0.0000

Data Index: 367933 (C4 (Web Text))

Max Activating Token Index: 10