Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 4038 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.6982. Min Act: -0.0000

Data Index: 200894 (C4 (Web Text))

Max Activating Token Index: 736

Text #1

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.6447. Min Act: -0.0000

Data Index: 767640 (C4 (Web Text))

Max Activating Token Index: 580

Text #2

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.6025. Min Act: -0.0000

Data Index: 1267498 (C4 (Web Text))

Max Activating Token Index: 552

Text #3

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.5159. Min Act: -0.0000

Data Index: 394237 (C4 (Web Text))

Max Activating Token Index: 937

Text #4

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4963. Min Act: -0.0000

Data Index: 473202 (C4 (Web Text))

Max Activating Token Index: 90

Text #5

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.5097. Min Act: -0.0000

Data Index: 231985 (C4 (Web Text))

Max Activating Token Index: 481

Text #6

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4826. Min Act: -0.0001

Data Index: 1260495 (C4 (Web Text))

Max Activating Token Index: 208

Text #7

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4749. Min Act: -0.0000

Data Index: 44953 (C4 (Web Text))

Max Activating Token Index: 70

Text #8

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4649. Min Act: 0.0001

Data Index: 1319167 (C4 (Web Text))

Max Activating Token Index: 853

Text #9

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4841. Min Act: 0.0000

Data Index: 479600 (C4 (Web Text))

Max Activating Token Index: 997

Text #10

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4672. Min Act: -0.0000

Data Index: 581305 (C4 (Web Text))

Max Activating Token Index: 49

Text #11

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4642. Min Act: 0.0000

Data Index: 535163 (C4 (Web Text))

Max Activating Token Index: 770

Text #12

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4265. Min Act: -0.0001

Data Index: 499698 (C4 (Web Text))

Max Activating Token Index: 153

Text #13

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4691. Min Act: -0.0000

Data Index: 1147540 (C4 (Web Text))

Max Activating Token Index: 418

Text #14

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4428. Min Act: -0.0000

Data Index: 1194735 (C4 (Web Text))

Max Activating Token Index: 945

Text #15

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4330. Min Act: 0.0001

Data Index: 169259 (C4 (Web Text))

Max Activating Token Index: 809

Text #16

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4462. Min Act: -0.0000

Data Index: 1019818 (C4 (Web Text))

Max Activating Token Index: 655

Text #17

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4346. Min Act: -0.0001

Data Index: 45360 (C4 (Web Text))

Max Activating Token Index: 545

Text #18

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4195. Min Act: 0.0001

Data Index: 943532 (C4 (Web Text))

Max Activating Token Index: 94

Text #19

Max Range: 0.6982. Min Range: -0.6982

Max Act: 0.4225. Min Act: -0.0001

Data Index: 1141130 (C4 (Web Text))

Max Activating Token Index: 787