Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2476 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1587. Min Act: -0.0001

Data Index: 14409 (C4 (Web Text))

Max Activating Token Index: 606

Text #1

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1469. Min Act: -0.0001

Data Index: 1215033 (C4 (Web Text))

Max Activating Token Index: 551

Text #2

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1404. Min Act: -0.0000

Data Index: 158793 (C4 (Web Text))

Max Activating Token Index: 766

Text #3

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1427. Min Act: -0.0001

Data Index: 671494 (C4 (Web Text))

Max Activating Token Index: 949

Text #4

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1380. Min Act: -0.0001

Data Index: 10536 (C4 (Web Text))

Max Activating Token Index: 988

Text #5

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1315. Min Act: -0.0001

Data Index: 764876 (C4 (Web Text))

Max Activating Token Index: 836

Text #6

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1290. Min Act: -0.0001

Data Index: 484290 (C4 (Web Text))

Max Activating Token Index: 817

Text #7

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1337. Min Act: -0.0001

Data Index: 718912 (C4 (Web Text))

Max Activating Token Index: 782

Text #8

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1283. Min Act: -0.0001

Data Index: 54439 (C4 (Web Text))

Max Activating Token Index: 35

Text #9

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1254. Min Act: -0.0001

Data Index: 883102 (C4 (Web Text))

Max Activating Token Index: 101

Text #10

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1238. Min Act: -0.0001

Data Index: 753658 (C4 (Web Text))

Max Activating Token Index: 508

Text #11

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1197. Min Act: -0.0001

Data Index: 121797 (C4 (Web Text))

Max Activating Token Index: 208

Text #12

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1233. Min Act: -0.0001

Data Index: 790721 (C4 (Web Text))

Max Activating Token Index: 481

Text #13

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1238. Min Act: -0.0000

Data Index: 621923 (C4 (Web Text))

Max Activating Token Index: 861

Text #14

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1174. Min Act: -0.0001

Data Index: 982980 (C4 (Web Text))

Max Activating Token Index: 424

Text #15

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1195. Min Act: -0.0000

Data Index: 1058389 (C4 (Web Text))

Max Activating Token Index: 830

Text #16

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1231. Min Act: -0.0000

Data Index: 927076 (C4 (Web Text))

Max Activating Token Index: 213

Text #17

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1091. Min Act: -0.0001

Data Index: 1276425 (C4 (Web Text))

Max Activating Token Index: 211

Text #18

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1175. Min Act: -0.0001

Data Index: 1031774 (C4 (Web Text))

Max Activating Token Index: 937

Text #19

Max Range: 0.1587. Min Range: -0.1587

Max Act: 0.1128. Min Act: -0.0001

Data Index: 1199624 (C4 (Web Text))

Max Activating Token Index: 464