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

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

Neuron 3780 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Click the toggle to see the full text!

Text #0

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9842. Min Act: -0.0002

Data Index: 844622 (C4 (Web Text))

Max Activating Token Index: 265

Text #1

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9786. Min Act: -0.0002

Data Index: 1501255 (Python Code)

Max Activating Token Index: 288

Text #2

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9542. Min Act: -0.0002

Data Index: 426224 (C4 (Web Text))

Max Activating Token Index: 556

Text #3

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9766. Min Act: -0.0002

Data Index: 834068 (C4 (Web Text))

Max Activating Token Index: 131

Text #4

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9474. Min Act: -0.0002

Data Index: 159634 (C4 (Web Text))

Max Activating Token Index: 41

Text #5

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9336. Min Act: -0.0002

Data Index: 30524 (C4 (Web Text))

Max Activating Token Index: 278

Text #6

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9453. Min Act: -0.0002

Data Index: 774261 (C4 (Web Text))

Max Activating Token Index: 162

Text #7

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9534. Min Act: -0.0002

Data Index: 926192 (C4 (Web Text))

Max Activating Token Index: 597

Text #8

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9183. Min Act: -0.0002

Data Index: 1574794 (Python Code)

Max Activating Token Index: 658

Text #9

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9308. Min Act: -0.0002

Data Index: 949931 (C4 (Web Text))

Max Activating Token Index: 256

Text #10

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9054. Min Act: -0.0002

Data Index: 710132 (C4 (Web Text))

Max Activating Token Index: 577

Text #11

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9178. Min Act: -0.0002

Data Index: 1272264 (C4 (Web Text))

Max Activating Token Index: 29

Text #12

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9070. Min Act: -0.0002

Data Index: 775961 (C4 (Web Text))

Max Activating Token Index: 578

Text #13

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9168. Min Act: -0.0002

Data Index: 1465387 (Python Code)

Max Activating Token Index: 340

Text #14

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9000. Min Act: -0.0002

Data Index: 661967 (C4 (Web Text))

Max Activating Token Index: 472

Text #15

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9085. Min Act: -0.0002

Data Index: 925055 (C4 (Web Text))

Max Activating Token Index: 827

Text #16

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.8957. Min Act: -0.0002

Data Index: 1389366 (Python Code)

Max Activating Token Index: 131

Text #17

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9055. Min Act: -0.0002

Data Index: 1439808 (Python Code)

Max Activating Token Index: 534

Text #18

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.9031. Min Act: -0.0002

Data Index: 981841 (C4 (Web Text))

Max Activating Token Index: 885

Text #19

Max Range: 0.9842. Min Range: -0.9842

Max Act: 0.8756. Min Act: -0.0002

Data Index: 853430 (C4 (Web Text))

Max Activating Token Index: 691