Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2417 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.3237. Min Act: -0.0001

Data Index: 1427542 (Python Code)

Max Activating Token Index: 42

Text #1

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.2893. Min Act: -0.0001

Data Index: 21048 (C4 (Web Text))

Max Activating Token Index: 176

Text #2

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.2458. Min Act: -0.0001

Data Index: 825533 (C4 (Web Text))

Max Activating Token Index: 161

Text #3

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.2926. Min Act: -0.0001

Data Index: 1344065 (C4 (Web Text))

Max Activating Token Index: 825

Text #4

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.2368. Min Act: -0.0001

Data Index: 1627531 (Python Code)

Max Activating Token Index: 17

Text #5

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.2216. Min Act: -0.0001

Data Index: 1117286 (C4 (Web Text))

Max Activating Token Index: 874

Text #6

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.2180. Min Act: -0.0001

Data Index: 1554964 (Python Code)

Max Activating Token Index: 40

Text #7

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.2123. Min Act: -0.0001

Data Index: 1405522 (Python Code)

Max Activating Token Index: 42

Text #8

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.1961. Min Act: -0.0001

Data Index: 236454 (C4 (Web Text))

Max Activating Token Index: 49

Text #9

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.2217. Min Act: -0.0001

Data Index: 371365 (C4 (Web Text))

Max Activating Token Index: 369

Text #10

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.1802. Min Act: -0.0001

Data Index: 1566785 (Python Code)

Max Activating Token Index: 14

Text #11

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.1095. Min Act: -0.0001

Data Index: 1588526 (Python Code)

Max Activating Token Index: 14

Text #12

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.1342. Min Act: -0.0001

Data Index: 1143742 (C4 (Web Text))

Max Activating Token Index: 36

Text #13

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.0923. Min Act: -0.0001

Data Index: 1538557 (Python Code)

Max Activating Token Index: 42

Text #14

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.1249. Min Act: -0.0001

Data Index: 275051 (C4 (Web Text))

Max Activating Token Index: 115

Text #15

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.1559. Min Act: -0.0001

Data Index: 1342888 (C4 (Web Text))

Max Activating Token Index: 311

Text #16

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.1459. Min Act: -0.0001

Data Index: 1240280 (C4 (Web Text))

Max Activating Token Index: 81

Text #17

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.0988. Min Act: -0.0001

Data Index: 1607941 (Python Code)

Max Activating Token Index: 404

Text #18

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.1240. Min Act: -0.0001

Data Index: 1513175 (Python Code)

Max Activating Token Index: 32

Text #19

Max Range: 1.3237. Min Range: -1.3237

Max Act: 1.0518. Min Act: -0.0001

Data Index: 1547289 (Python Code)

Max Activating Token Index: 88