Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3822 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.6777. Min Act: -0.0000

Data Index: 434631 (C4 (Web Text))

Max Activating Token Index: 237

Text #1

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.6929. Min Act: -0.0000

Data Index: 290413 (C4 (Web Text))

Max Activating Token Index: 20

Text #2

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.6365. Min Act: -0.0000

Data Index: 1605323 (Python Code)

Max Activating Token Index: 955

Text #3

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.4826. Min Act: -0.0000

Data Index: 644047 (C4 (Web Text))

Max Activating Token Index: 322

Text #4

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.5658. Min Act: -0.0000

Data Index: 1114406 (C4 (Web Text))

Max Activating Token Index: 496

Text #5

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.5535. Min Act: -0.0000

Data Index: 161023 (C4 (Web Text))

Max Activating Token Index: 710

Text #6

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.5960. Min Act: -0.0000

Data Index: 1461978 (Python Code)

Max Activating Token Index: 279

Text #7

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.4181. Min Act: -0.0000

Data Index: 226873 (C4 (Web Text))

Max Activating Token Index: 421

Text #8

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.4939. Min Act: -0.0000

Data Index: 353272 (C4 (Web Text))

Max Activating Token Index: 431

Text #9

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.4957. Min Act: -0.0000

Data Index: 188582 (C4 (Web Text))

Max Activating Token Index: 32

Text #10

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.4775. Min Act: -0.0000

Data Index: 61433 (C4 (Web Text))

Max Activating Token Index: 612

Text #11

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.4614. Min Act: -0.0000

Data Index: 1534137 (Python Code)

Max Activating Token Index: 708

Text #12

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.4658. Min Act: -0.0000

Data Index: 894614 (C4 (Web Text))

Max Activating Token Index: 840

Text #13

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.4465. Min Act: -0.0000

Data Index: 1078800 (C4 (Web Text))

Max Activating Token Index: 682

Text #14

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.4179. Min Act: -0.0000

Data Index: 95687 (C4 (Web Text))

Max Activating Token Index: 154

Text #15

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.4268. Min Act: -0.0000

Data Index: 1271961 (C4 (Web Text))

Max Activating Token Index: 968

Text #16

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.4237. Min Act: -0.0000

Data Index: 606363 (C4 (Web Text))

Max Activating Token Index: 720

Text #17

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.4993. Min Act: -0.0000

Data Index: 1239554 (C4 (Web Text))

Max Activating Token Index: 49

Text #18

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.3202. Min Act: -0.0000

Data Index: 590142 (C4 (Web Text))

Max Activating Token Index: 944

Text #19

Max Range: 1.6929. Min Range: -1.6929

Max Act: 1.3867. Min Act: -0.0000

Data Index: 1339799 (C4 (Web Text))

Max Activating Token Index: 764