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

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

Neuron 772 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.5699. Min Act: -0.0000

Data Index: 1103426 (C4 (Web Text))

Max Activating Token Index: 926

Text #1

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.5440. Min Act: -0.0000

Data Index: 707218 (C4 (Web Text))

Max Activating Token Index: 984

Text #2

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.6311. Min Act: -0.0000

Data Index: 600817 (C4 (Web Text))

Max Activating Token Index: 710

Text #3

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.5750. Min Act: -0.0000

Data Index: 257495 (C4 (Web Text))

Max Activating Token Index: 611

Text #4

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.4293. Min Act: -0.0000

Data Index: 734230 (C4 (Web Text))

Max Activating Token Index: 150

Text #5

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.5076. Min Act: -0.0000

Data Index: 1525894 (Python Code)

Max Activating Token Index: 526

Text #6

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.4269. Min Act: -0.0000

Data Index: 326038 (C4 (Web Text))

Max Activating Token Index: 197

Text #7

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.4898. Min Act: -0.0000

Data Index: 1531329 (Python Code)

Max Activating Token Index: 924

Text #8

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.4472. Min Act: -0.0000

Data Index: 681076 (C4 (Web Text))

Max Activating Token Index: 51

Text #9

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.4110. Min Act: -0.0000

Data Index: 1364002 (Python Code)

Max Activating Token Index: 872

Text #10

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.4380. Min Act: -0.0000

Data Index: 45000 (C4 (Web Text))

Max Activating Token Index: 55

Text #11

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.3133. Min Act: -0.0000

Data Index: 1373371 (Python Code)

Max Activating Token Index: 895

Text #12

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.4674. Min Act: -0.0000

Data Index: 1150518 (C4 (Web Text))

Max Activating Token Index: 234

Text #13

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.4179. Min Act: -0.0000

Data Index: 278920 (C4 (Web Text))

Max Activating Token Index: 846

Text #14

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.3004. Min Act: -0.0000

Data Index: 815313 (C4 (Web Text))

Max Activating Token Index: 84

Text #15

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.3629. Min Act: -0.0000

Data Index: 1482220 (Python Code)

Max Activating Token Index: 962

Text #16

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.2869. Min Act: -0.0000

Data Index: 1030720 (C4 (Web Text))

Max Activating Token Index: 269

Text #17

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.3601. Min Act: -0.0000

Data Index: 213060 (C4 (Web Text))

Max Activating Token Index: 350

Text #18

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.3516. Min Act: -0.0000

Data Index: 298006 (C4 (Web Text))

Max Activating Token Index: 257

Text #19

Max Range: 1.6311. Min Range: -1.6311

Max Act: 1.2307. Min Act: -0.0000

Data Index: 1240974 (C4 (Web Text))

Max Activating Token Index: 772