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

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

Neuron 3574 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0340. Min Act: -0.0001

Data Index: 242142 (C4 (Web Text))

Max Activating Token Index: 963

Text #1

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0368. Min Act: -0.0001

Data Index: 1590871 (Python Code)

Max Activating Token Index: 472

Text #2

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0343. Min Act: -0.0001

Data Index: 1590341 (Python Code)

Max Activating Token Index: 257

Text #3

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0315. Min Act: -0.0001

Data Index: 237532 (C4 (Web Text))

Max Activating Token Index: 633

Text #4

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0310. Min Act: -0.0001

Data Index: 999020 (C4 (Web Text))

Max Activating Token Index: 444

Text #5

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0343. Min Act: -0.0001

Data Index: 1069006 (C4 (Web Text))

Max Activating Token Index: 935

Text #6

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0343. Min Act: -0.0001

Data Index: 737876 (C4 (Web Text))

Max Activating Token Index: 124

Text #7

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0314. Min Act: -0.0001

Data Index: 885093 (C4 (Web Text))

Max Activating Token Index: 150

Text #8

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0311. Min Act: -0.0001

Data Index: 113627 (C4 (Web Text))

Max Activating Token Index: 870

Text #9

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0312. Min Act: -0.0001

Data Index: 129457 (C4 (Web Text))

Max Activating Token Index: 483

Text #10

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0304. Min Act: -0.0001

Data Index: 1528699 (Python Code)

Max Activating Token Index: 289

Text #11

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0288. Min Act: -0.0001

Data Index: 1627487 (Python Code)

Max Activating Token Index: 554

Text #12

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0284. Min Act: -0.0001

Data Index: 73740 (C4 (Web Text))

Max Activating Token Index: 295

Text #13

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0307. Min Act: -0.0001

Data Index: 557723 (C4 (Web Text))

Max Activating Token Index: 144

Text #14

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0295. Min Act: -0.0001

Data Index: 750596 (C4 (Web Text))

Max Activating Token Index: 410

Text #15

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0305. Min Act: -0.0001

Data Index: 1166671 (C4 (Web Text))

Max Activating Token Index: 879

Text #16

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0278. Min Act: -0.0001

Data Index: 1268763 (C4 (Web Text))

Max Activating Token Index: 648

Text #17

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0292. Min Act: -0.0001

Data Index: 708866 (C4 (Web Text))

Max Activating Token Index: 74

Text #18

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0311. Min Act: -0.0001

Data Index: 394398 (C4 (Web Text))

Max Activating Token Index: 753

Text #19

Max Range: 0.0368. Min Range: -0.0368

Max Act: 0.0275. Min Act: -0.0001

Data Index: 566425 (C4 (Web Text))

Max Activating Token Index: 750