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

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

Neuron 1002 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1169. Min Range: -1.1169

Max Act: 1.1169. Min Act: -0.0001

Data Index: 408091 (C4 (Web Text))

Max Activating Token Index: 1008

Text #1

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.8374. Min Act: -0.0001

Data Index: 1545608 (Python Code)

Max Activating Token Index: 348

Text #2

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.9101. Min Act: -0.0001

Data Index: 1559093 (Python Code)

Max Activating Token Index: 140

Text #3

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.9133. Min Act: -0.0001

Data Index: 641467 (C4 (Web Text))

Max Activating Token Index: 114

Text #4

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.8298. Min Act: -0.0001

Data Index: 522786 (C4 (Web Text))

Max Activating Token Index: 192

Text #5

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.8220. Min Act: -0.0001

Data Index: 1420935 (Python Code)

Max Activating Token Index: 545

Text #6

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.8346. Min Act: -0.0001

Data Index: 1599528 (Python Code)

Max Activating Token Index: 184

Text #7

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.8408. Min Act: -0.0001

Data Index: 1568900 (Python Code)

Max Activating Token Index: 71

Text #8

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.8151. Min Act: -0.0001

Data Index: 1365985 (Python Code)

Max Activating Token Index: 423

Text #9

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.8287. Min Act: -0.0001

Data Index: 1636848 (Python Code)

Max Activating Token Index: 27

Text #10

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.8060. Min Act: -0.0001

Data Index: 28709 (C4 (Web Text))

Max Activating Token Index: 43

Text #11

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.7477. Min Act: -0.0001

Data Index: 1397990 (Python Code)

Max Activating Token Index: 808

Text #12

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.8424. Min Act: -0.0001

Data Index: 1384568 (Python Code)

Max Activating Token Index: 394

Text #13

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.8084. Min Act: -0.0001

Data Index: 566823 (C4 (Web Text))

Max Activating Token Index: 829

Text #14

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.7487. Min Act: -0.0001

Data Index: 1544024 (Python Code)

Max Activating Token Index: 50

Text #15

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.7564. Min Act: -0.0001

Data Index: 1608154 (Python Code)

Max Activating Token Index: 793

Text #16

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.7877. Min Act: -0.0001

Data Index: 1404817 (Python Code)

Max Activating Token Index: 854

Text #17

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.8030. Min Act: -0.0001

Data Index: 531528 (C4 (Web Text))

Max Activating Token Index: 916

Text #18

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.6944. Min Act: -0.0001

Data Index: 846208 (C4 (Web Text))

Max Activating Token Index: 802

Text #19

Max Range: 1.1169. Min Range: -1.1169

Max Act: 0.7184. Min Act: -0.0001

Data Index: 790438 (C4 (Web Text))

Max Activating Token Index: 176