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

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

Neuron 5671 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.8763. Min Act: -0.0002

Data Index: 1149644 (C4 (Web Text))

Max Activating Token Index: 384

Text #1

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.5762. Min Act: -0.0001

Data Index: 1387795 (Python Code)

Max Activating Token Index: 272

Text #2

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.5839. Min Act: -0.0001

Data Index: 392433 (C4 (Web Text))

Max Activating Token Index: 540

Text #3

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.4926. Min Act: -0.0001

Data Index: 1582766 (Python Code)

Max Activating Token Index: 44

Text #4

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.5286. Min Act: -0.0001

Data Index: 489809 (C4 (Web Text))

Max Activating Token Index: 462

Text #5

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.4492. Min Act: -0.0001

Data Index: 1589300 (Python Code)

Max Activating Token Index: 101

Text #6

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.3349. Min Act: -0.0001

Data Index: 1525285 (Python Code)

Max Activating Token Index: 208

Text #7

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.3411. Min Act: -0.0001

Data Index: 850110 (C4 (Web Text))

Max Activating Token Index: 686

Text #8

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.6557. Min Act: -0.0001

Data Index: 1253604 (C4 (Web Text))

Max Activating Token Index: 64

Text #9

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.3243. Min Act: -0.0001

Data Index: 1379573 (Python Code)

Max Activating Token Index: 123

Text #10

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.4798. Min Act: -0.0001

Data Index: 577827 (C4 (Web Text))

Max Activating Token Index: 238

Text #11

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.3247. Min Act: -0.0001

Data Index: 1646253 (Python Code)

Max Activating Token Index: 245

Text #12

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.3502. Min Act: -0.0001

Data Index: 496009 (C4 (Web Text))

Max Activating Token Index: 912

Text #13

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.4819. Min Act: -0.0001

Data Index: 1610569 (Python Code)

Max Activating Token Index: 105

Text #14

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.4845. Min Act: -0.0001

Data Index: 973187 (C4 (Web Text))

Max Activating Token Index: 996

Text #15

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.4394. Min Act: -0.0001

Data Index: 66059 (C4 (Web Text))

Max Activating Token Index: 191

Text #16

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.4200. Min Act: -0.0001

Data Index: 1117763 (C4 (Web Text))

Max Activating Token Index: 101

Text #17

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.1962. Min Act: -0.0001

Data Index: 381874 (C4 (Web Text))

Max Activating Token Index: 372

Text #18

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.4028. Min Act: -0.0001

Data Index: 1369338 (Python Code)

Max Activating Token Index: 43

Text #19

Max Range: 1.8763. Min Range: -1.8763

Max Act: 1.5241. Min Act: -0.0001

Data Index: 924753 (C4 (Web Text))

Max Activating Token Index: 523