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

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

Neuron 5424 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0408. Min Act: -0.0001

Data Index: 1058786 (C4 (Web Text))

Max Activating Token Index: 239

Text #1

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0351. Min Act: -0.0001

Data Index: 648085 (C4 (Web Text))

Max Activating Token Index: 673

Text #2

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0313. Min Act: -0.0001

Data Index: 1383364 (Python Code)

Max Activating Token Index: 259

Text #3

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0324. Min Act: -0.0001

Data Index: 1313139 (C4 (Web Text))

Max Activating Token Index: 4

Text #4

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0304. Min Act: -0.0001

Data Index: 754581 (C4 (Web Text))

Max Activating Token Index: 14

Text #5

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0315. Min Act: -0.0001

Data Index: 1246174 (C4 (Web Text))

Max Activating Token Index: 23

Text #6

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0320. Min Act: -0.0001

Data Index: 910471 (C4 (Web Text))

Max Activating Token Index: 9

Text #7

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0292. Min Act: -0.0001

Data Index: 242086 (C4 (Web Text))

Max Activating Token Index: 569

Text #8

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0293. Min Act: -0.0001

Data Index: 1042644 (C4 (Web Text))

Max Activating Token Index: 931

Text #9

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0290. Min Act: -0.0001

Data Index: 1052745 (C4 (Web Text))

Max Activating Token Index: 94

Text #10

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0277. Min Act: -0.0001

Data Index: 909213 (C4 (Web Text))

Max Activating Token Index: 591

Text #11

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0281. Min Act: -0.0001

Data Index: 1443376 (Python Code)

Max Activating Token Index: 776

Text #12

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0283. Min Act: -0.0001

Data Index: 812513 (C4 (Web Text))

Max Activating Token Index: 688

Text #13

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0283. Min Act: -0.0001

Data Index: 829536 (C4 (Web Text))

Max Activating Token Index: 825

Text #14

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0271. Min Act: -0.0001

Data Index: 1261131 (C4 (Web Text))

Max Activating Token Index: 226

Text #15

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0258. Min Act: -0.0001

Data Index: 1239046 (C4 (Web Text))

Max Activating Token Index: 612

Text #16

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0268. Min Act: -0.0001

Data Index: 1173596 (C4 (Web Text))

Max Activating Token Index: 804

Text #17

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0293. Min Act: -0.0001

Data Index: 990642 (C4 (Web Text))

Max Activating Token Index: 151

Text #18

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0276. Min Act: -0.0001

Data Index: 687914 (C4 (Web Text))

Max Activating Token Index: 307

Text #19

Max Range: 0.0408. Min Range: -0.0408

Max Act: 0.0258. Min Act: -0.0001

Data Index: 970693 (C4 (Web Text))

Max Activating Token Index: 442