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

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

Neuron 4558 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.7640. Min Act: -0.0000

Data Index: 208504 (C4 (Web Text))

Max Activating Token Index: 539

Text #1

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.7614. Min Act: -0.0000

Data Index: 1097640 (C4 (Web Text))

Max Activating Token Index: 822

Text #2

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.8210. Min Act: -0.0000

Data Index: 499140 (C4 (Web Text))

Max Activating Token Index: 277

Text #3

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.5559. Min Act: -0.0000

Data Index: 71572 (C4 (Web Text))

Max Activating Token Index: 35

Text #4

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.3528. Min Act: -0.0000

Data Index: 746553 (C4 (Web Text))

Max Activating Token Index: 946

Text #5

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.4322. Min Act: -0.0000

Data Index: 1582513 (Python Code)

Max Activating Token Index: 201

Text #6

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.6817. Min Act: -0.0000

Data Index: 851979 (C4 (Web Text))

Max Activating Token Index: 724

Text #7

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.2904. Min Act: -0.0000

Data Index: 93805 (C4 (Web Text))

Max Activating Token Index: 410

Text #8

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.4762. Min Act: -0.0000

Data Index: 252335 (C4 (Web Text))

Max Activating Token Index: 147

Text #9

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.3570. Min Act: -0.0000

Data Index: 15037 (C4 (Web Text))

Max Activating Token Index: 187

Text #10

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.2138. Min Act: -0.0000

Data Index: 587473 (C4 (Web Text))

Max Activating Token Index: 289

Text #11

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.2932. Min Act: -0.0000

Data Index: 851918 (C4 (Web Text))

Max Activating Token Index: 158

Text #12

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.1580. Min Act: -0.0000

Data Index: 202683 (C4 (Web Text))

Max Activating Token Index: 28

Text #13

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.2627. Min Act: -0.0000

Data Index: 118389 (C4 (Web Text))

Max Activating Token Index: 762

Text #14

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.2322. Min Act: -0.0000

Data Index: 279523 (C4 (Web Text))

Max Activating Token Index: 493

Text #15

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.2354. Min Act: -0.0000

Data Index: 220623 (C4 (Web Text))

Max Activating Token Index: 128

Text #16

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.2331. Min Act: -0.0000

Data Index: 1171968 (C4 (Web Text))

Max Activating Token Index: 311

Text #17

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.2008. Min Act: -0.0000

Data Index: 1118660 (C4 (Web Text))

Max Activating Token Index: 775

Text #18

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.1965. Min Act: -0.0000

Data Index: 1618309 (Python Code)

Max Activating Token Index: 51

Text #19

Max Range: 2.8210. Min Range: -2.8210

Max Act: 2.0653. Min Act: -0.0000

Data Index: 647291 (C4 (Web Text))

Max Activating Token Index: 490