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

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

Neuron 1197 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.8223. Min Act: -0.0003

Data Index: 914841 (C4 (Web Text))

Max Activating Token Index: 540

Text #1

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.7936. Min Act: -0.0003

Data Index: 1000562 (C4 (Web Text))

Max Activating Token Index: 316

Text #2

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.7817. Min Act: -0.0003

Data Index: 290939 (C4 (Web Text))

Max Activating Token Index: 117

Text #3

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.7720. Min Act: -0.0003

Data Index: 856067 (C4 (Web Text))

Max Activating Token Index: 529

Text #4

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.7471. Min Act: -0.0003

Data Index: 929322 (C4 (Web Text))

Max Activating Token Index: 592

Text #5

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.6728. Min Act: -0.0003

Data Index: 1017076 (C4 (Web Text))

Max Activating Token Index: 937

Text #6

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.6607. Min Act: -0.0003

Data Index: 628685 (C4 (Web Text))

Max Activating Token Index: 313

Text #7

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.6335. Min Act: -0.0003

Data Index: 888396 (C4 (Web Text))

Max Activating Token Index: 596

Text #8

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.6560. Min Act: -0.0003

Data Index: 1038168 (C4 (Web Text))

Max Activating Token Index: 280

Text #9

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.6262. Min Act: -0.0003

Data Index: 1112959 (C4 (Web Text))

Max Activating Token Index: 511

Text #10

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.6745. Min Act: -0.0003

Data Index: 802853 (C4 (Web Text))

Max Activating Token Index: 778

Text #11

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.7334. Min Act: -0.0003

Data Index: 1324420 (C4 (Web Text))

Max Activating Token Index: 929

Text #12

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.6573. Min Act: -0.0003

Data Index: 1341824 (C4 (Web Text))

Max Activating Token Index: 74

Text #13

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.6062. Min Act: -0.0003

Data Index: 688329 (C4 (Web Text))

Max Activating Token Index: 1004

Text #14

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.5997. Min Act: -0.0003

Data Index: 1363022 (Python Code)

Max Activating Token Index: 473

Text #15

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.6521. Min Act: -0.0003

Data Index: 310019 (C4 (Web Text))

Max Activating Token Index: 649

Text #16

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.5204. Min Act: -0.0003

Data Index: 1061714 (C4 (Web Text))

Max Activating Token Index: 178

Text #17

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.6280. Min Act: -0.0003

Data Index: 1477908 (Python Code)

Max Activating Token Index: 121

Text #18

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.5816. Min Act: -0.0003

Data Index: 290472 (C4 (Web Text))

Max Activating Token Index: 310

Text #19

Max Range: 0.8223. Min Range: -0.8223

Max Act: 0.6218. Min Act: -0.0003

Data Index: 150000 (C4 (Web Text))

Max Activating Token Index: 867