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

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

Neuron 3157 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.2169. Min Range: -1.2169

Max Act: 1.2169. Min Act: -0.0000

Data Index: 337712 (C4 (Web Text))

Max Activating Token Index: 597

Text #1

Max Range: 1.2169. Min Range: -1.2169

Max Act: 1.1185. Min Act: -0.0000

Data Index: 1572350 (Python Code)

Max Activating Token Index: 295

Text #2

Max Range: 1.2169. Min Range: -1.2169

Max Act: 1.0297. Min Act: -0.0000

Data Index: 1423824 (Python Code)

Max Activating Token Index: 684

Text #3

Max Range: 1.2169. Min Range: -1.2169

Max Act: 1.0124. Min Act: -0.0000

Data Index: 732054 (C4 (Web Text))

Max Activating Token Index: 753

Text #4

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.9743. Min Act: -0.0000

Data Index: 1318739 (C4 (Web Text))

Max Activating Token Index: 853

Text #5

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.9550. Min Act: -0.0000

Data Index: 1558575 (Python Code)

Max Activating Token Index: 858

Text #6

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.9227. Min Act: -0.0000

Data Index: 1551615 (Python Code)

Max Activating Token Index: 52

Text #7

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.9111. Min Act: -0.0000

Data Index: 1407962 (Python Code)

Max Activating Token Index: 729

Text #8

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.8537. Min Act: -0.0000

Data Index: 1353300 (C4 (Web Text))

Max Activating Token Index: 302

Text #9

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.8577. Min Act: -0.0000

Data Index: 368324 (C4 (Web Text))

Max Activating Token Index: 961

Text #10

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.8638. Min Act: -0.0000

Data Index: 977532 (C4 (Web Text))

Max Activating Token Index: 999

Text #11

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.8091. Min Act: -0.0000

Data Index: 596218 (C4 (Web Text))

Max Activating Token Index: 563

Text #12

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.8710. Min Act: -0.0000

Data Index: 1452102 (Python Code)

Max Activating Token Index: 188

Text #13

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.8208. Min Act: -0.0000

Data Index: 1243551 (C4 (Web Text))

Max Activating Token Index: 738

Text #14

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.8505. Min Act: -0.0000

Data Index: 1125163 (C4 (Web Text))

Max Activating Token Index: 143

Text #15

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.7789. Min Act: -0.0000

Data Index: 1626516 (Python Code)

Max Activating Token Index: 42

Text #16

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.8024. Min Act: -0.0000

Data Index: 1650270 (Python Code)

Max Activating Token Index: 826

Text #17

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.7967. Min Act: -0.0000

Data Index: 1134753 (C4 (Web Text))

Max Activating Token Index: 1001

Text #18

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.7663. Min Act: -0.0000

Data Index: 1467355 (Python Code)

Max Activating Token Index: 469

Text #19

Max Range: 1.2169. Min Range: -1.2169

Max Act: 0.7745. Min Act: -0.0000

Data Index: 581009 (C4 (Web Text))

Max Activating Token Index: 1018