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

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

Neuron 4022 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.8977. Min Act: -0.0001

Data Index: 166401 (C4 (Web Text))

Max Activating Token Index: 693

Text #1

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.8324. Min Act: -0.0001

Data Index: 427536 (C4 (Web Text))

Max Activating Token Index: 704

Text #2

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.4296. Min Act: -0.0001

Data Index: 1592456 (Python Code)

Max Activating Token Index: 47

Text #3

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.5395. Min Act: -0.0001

Data Index: 1001196 (C4 (Web Text))

Max Activating Token Index: 32

Text #4

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.5786. Min Act: -0.0001

Data Index: 1382979 (Python Code)

Max Activating Token Index: 826

Text #5

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.4692. Min Act: -0.0001

Data Index: 700253 (C4 (Web Text))

Max Activating Token Index: 111

Text #6

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.3975. Min Act: -0.0001

Data Index: 1603711 (Python Code)

Max Activating Token Index: 301

Text #7

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.3527. Min Act: -0.0001

Data Index: 186603 (C4 (Web Text))

Max Activating Token Index: 915

Text #8

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.2525. Min Act: -0.0001

Data Index: 1191345 (C4 (Web Text))

Max Activating Token Index: 229

Text #9

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.3994. Min Act: -0.0001

Data Index: 161355 (C4 (Web Text))

Max Activating Token Index: 570

Text #10

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.3868. Min Act: -0.0001

Data Index: 348940 (C4 (Web Text))

Max Activating Token Index: 816

Text #11

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.3175. Min Act: -0.0001

Data Index: 238755 (C4 (Web Text))

Max Activating Token Index: 824

Text #12

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.3393. Min Act: -0.0001

Data Index: 1241314 (C4 (Web Text))

Max Activating Token Index: 1016

Text #13

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.2904. Min Act: -0.0001

Data Index: 954097 (C4 (Web Text))

Max Activating Token Index: 588

Text #14

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.3123. Min Act: -0.0001

Data Index: 1285730 (C4 (Web Text))

Max Activating Token Index: 699

Text #15

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.3205. Min Act: -0.0001

Data Index: 968527 (C4 (Web Text))

Max Activating Token Index: 889

Text #16

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.2830. Min Act: -0.0001

Data Index: 947570 (C4 (Web Text))

Max Activating Token Index: 450

Text #17

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.1597. Min Act: -0.0001

Data Index: 1300844 (C4 (Web Text))

Max Activating Token Index: 531

Text #18

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.2614. Min Act: -0.0001

Data Index: 789070 (C4 (Web Text))

Max Activating Token Index: 480

Text #19

Max Range: 1.8977. Min Range: -1.8977

Max Act: 1.1898. Min Act: -0.0001

Data Index: 764696 (C4 (Web Text))

Max Activating Token Index: 180