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

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

Neuron 3456 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.6155. Min Range: -4.6155

Max Act: 4.6155. Min Act: -0.0000

Data Index: 55722 (C4 (Web Text))

Max Activating Token Index: 219

Text #1

Max Range: 4.6155. Min Range: -4.6155

Max Act: 4.3464. Min Act: -0.0000

Data Index: 360137 (C4 (Web Text))

Max Activating Token Index: 138

Text #2

Max Range: 4.6155. Min Range: -4.6155

Max Act: 4.3940. Min Act: -0.0000

Data Index: 680505 (C4 (Web Text))

Max Activating Token Index: 222

Text #3

Max Range: 4.6155. Min Range: -4.6155

Max Act: 3.3991. Min Act: -0.0000

Data Index: 978349 (C4 (Web Text))

Max Activating Token Index: 198

Text #4

Max Range: 4.6155. Min Range: -4.6155

Max Act: 3.6241. Min Act: -0.0000

Data Index: 1306811 (C4 (Web Text))

Max Activating Token Index: 75

Text #5

Max Range: 4.6155. Min Range: -4.6155

Max Act: 3.7174. Min Act: -0.0000

Data Index: 422147 (C4 (Web Text))

Max Activating Token Index: 437

Text #6

Max Range: 4.6155. Min Range: -4.6155

Max Act: 4.4244. Min Act: -0.0000

Data Index: 1150697 (C4 (Web Text))

Max Activating Token Index: 299

Text #7

Max Range: 4.6155. Min Range: -4.6155

Max Act: 3.2389. Min Act: -0.0000

Data Index: 851366 (C4 (Web Text))

Max Activating Token Index: 536

Text #8

Max Range: 4.6155. Min Range: -4.6155

Max Act: 2.3482. Min Act: -0.0000

Data Index: 400768 (C4 (Web Text))

Max Activating Token Index: 161

Text #9

Max Range: 4.6155. Min Range: -4.6155

Max Act: 2.6203. Min Act: -0.0000

Data Index: 1099950 (C4 (Web Text))

Max Activating Token Index: 471

Text #10

Max Range: 4.6155. Min Range: -4.6155

Max Act: 2.0691. Min Act: -0.0000

Data Index: 1223988 (C4 (Web Text))

Max Activating Token Index: 517

Text #11

Max Range: 4.6155. Min Range: -4.6155

Max Act: 1.6187. Min Act: -0.0000

Data Index: 26670 (C4 (Web Text))

Max Activating Token Index: 844

Text #12

Max Range: 4.6155. Min Range: -4.6155

Max Act: 1.8078. Min Act: -0.0000

Data Index: 204899 (C4 (Web Text))

Max Activating Token Index: 83

Text #13

Max Range: 4.6155. Min Range: -4.6155

Max Act: 2.5350. Min Act: -0.0000

Data Index: 332890 (C4 (Web Text))

Max Activating Token Index: 604

Text #14

Max Range: 4.6155. Min Range: -4.6155

Max Act: 2.0564. Min Act: -0.0000

Data Index: 1108751 (C4 (Web Text))

Max Activating Token Index: 417

Text #15

Max Range: 4.6155. Min Range: -4.6155

Max Act: 1.1171. Min Act: -0.0000

Data Index: 811725 (C4 (Web Text))

Max Activating Token Index: 868

Text #16

Max Range: 4.6155. Min Range: -4.6155

Max Act: 1.5027. Min Act: -0.0000

Data Index: 447246 (C4 (Web Text))

Max Activating Token Index: 614

Text #17

Max Range: 4.6155. Min Range: -4.6155

Max Act: 1.1214. Min Act: -0.0000

Data Index: 1448395 (Python Code)

Max Activating Token Index: 439

Text #18

Max Range: 4.6155. Min Range: -4.6155

Max Act: 1.2970. Min Act: -0.0000

Data Index: 209253 (C4 (Web Text))

Max Activating Token Index: 990

Text #19

Max Range: 4.6155. Min Range: -4.6155

Max Act: 1.0000. Min Act: -0.0000

Data Index: 391695 (C4 (Web Text))

Max Activating Token Index: 596