Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1205 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0332. Min Act: -0.0005

Data Index: 487711 (C4 (Web Text))

Max Activating Token Index: 289

Text #1

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0341. Min Act: -0.0005

Data Index: 689538 (C4 (Web Text))

Max Activating Token Index: 133

Text #2

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0327. Min Act: -0.0005

Data Index: 463861 (C4 (Web Text))

Max Activating Token Index: 290

Text #3

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0320. Min Act: -0.0005

Data Index: 93325 (C4 (Web Text))

Max Activating Token Index: 271

Text #4

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0314. Min Act: -0.0005

Data Index: 1471543 (Python Code)

Max Activating Token Index: 770

Text #5

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0306. Min Act: -0.0005

Data Index: 347906 (C4 (Web Text))

Max Activating Token Index: 170

Text #6

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0332. Min Act: -0.0005

Data Index: 1015684 (C4 (Web Text))

Max Activating Token Index: 333

Text #7

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0299. Min Act: -0.0005

Data Index: 1349246 (C4 (Web Text))

Max Activating Token Index: 454

Text #8

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0309. Min Act: -0.0005

Data Index: 1128120 (C4 (Web Text))

Max Activating Token Index: 244

Text #9

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0317. Min Act: -0.0005

Data Index: 1503714 (Python Code)

Max Activating Token Index: 864

Text #10

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0302. Min Act: -0.0005

Data Index: 1310429 (C4 (Web Text))

Max Activating Token Index: 334

Text #11

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0294. Min Act: -0.0005

Data Index: 1512335 (Python Code)

Max Activating Token Index: 353

Text #12

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0293. Min Act: -0.0005

Data Index: 1216727 (C4 (Web Text))

Max Activating Token Index: 900

Text #13

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0299. Min Act: -0.0005

Data Index: 876253 (C4 (Web Text))

Max Activating Token Index: 761

Text #14

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0298. Min Act: -0.0005

Data Index: 1342094 (C4 (Web Text))

Max Activating Token Index: 829

Text #15

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0304. Min Act: -0.0005

Data Index: 1554497 (Python Code)

Max Activating Token Index: 977

Text #16

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0292. Min Act: -0.0005

Data Index: 755042 (C4 (Web Text))

Max Activating Token Index: 492

Text #17

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0309. Min Act: -0.0005

Data Index: 1567218 (Python Code)

Max Activating Token Index: 825

Text #18

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0290. Min Act: -0.0005

Data Index: 1647958 (Python Code)

Max Activating Token Index: 672

Text #19

Max Range: 0.0341. Min Range: -0.0341

Max Act: 0.0295. Min Act: -0.0005

Data Index: 1634303 (Python Code)

Max Activating Token Index: 732