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

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

Neuron 4674 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.5314. Min Range: -1.5314

Max Act: 1.5314. Min Act: -0.0000

Data Index: 403537 (C4 (Web Text))

Max Activating Token Index: 834

Text #1

Max Range: 1.5314. Min Range: -1.5314

Max Act: 1.2749. Min Act: -0.0000

Data Index: 911309 (C4 (Web Text))

Max Activating Token Index: 117

Text #2

Max Range: 1.5314. Min Range: -1.5314

Max Act: 1.1940. Min Act: -0.0000

Data Index: 674459 (C4 (Web Text))

Max Activating Token Index: 1001

Text #3

Max Range: 1.5314. Min Range: -1.5314

Max Act: 1.1149. Min Act: -0.0000

Data Index: 324738 (C4 (Web Text))

Max Activating Token Index: 546

Text #4

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.8428. Min Act: -0.0000

Data Index: 243303 (C4 (Web Text))

Max Activating Token Index: 75

Text #5

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.7969. Min Act: -0.0000

Data Index: 521050 (C4 (Web Text))

Max Activating Token Index: 205

Text #6

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.7256. Min Act: -0.0000

Data Index: 670815 (C4 (Web Text))

Max Activating Token Index: 170

Text #7

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.6787. Min Act: -0.0000

Data Index: 328757 (C4 (Web Text))

Max Activating Token Index: 52

Text #8

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.6442. Min Act: -0.0000

Data Index: 740913 (C4 (Web Text))

Max Activating Token Index: 173

Text #9

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.6670. Min Act: -0.0000

Data Index: 1465675 (Python Code)

Max Activating Token Index: 90

Text #10

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.6227. Min Act: -0.0000

Data Index: 996467 (C4 (Web Text))

Max Activating Token Index: 671

Text #11

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.6144. Min Act: -0.0000

Data Index: 1263642 (C4 (Web Text))

Max Activating Token Index: 161

Text #12

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.5868. Min Act: -0.0000

Data Index: 190410 (C4 (Web Text))

Max Activating Token Index: 789

Text #13

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.5949. Min Act: -0.0000

Data Index: 1574644 (Python Code)

Max Activating Token Index: 99

Text #14

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.5713. Min Act: -0.0000

Data Index: 902732 (C4 (Web Text))

Max Activating Token Index: 592

Text #15

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.5671. Min Act: -0.0000

Data Index: 857897 (C4 (Web Text))

Max Activating Token Index: 621

Text #16

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.5787. Min Act: -0.0000

Data Index: 600515 (C4 (Web Text))

Max Activating Token Index: 389

Text #17

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.5589. Min Act: -0.0000

Data Index: 1655764 (Python Code)

Max Activating Token Index: 116

Text #18

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.5588. Min Act: -0.0000

Data Index: 1655168 (Python Code)

Max Activating Token Index: 85

Text #19

Max Range: 1.5314. Min Range: -1.5314

Max Act: 0.5590. Min Act: -0.0000

Data Index: 284205 (C4 (Web Text))

Max Activating Token Index: 473