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

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

Neuron 3676 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: 0.1883. Min Range: -0.1883

Max Act: 0.1649. Min Act: -0.0001

Data Index: 290856 (C4 (Web Text))

Max Activating Token Index: 304

Text #1

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1443. Min Act: -0.0001

Data Index: 838592 (C4 (Web Text))

Max Activating Token Index: 708

Text #2

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1883. Min Act: -0.0001

Data Index: 1112126 (C4 (Web Text))

Max Activating Token Index: 455

Text #3

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1132. Min Act: -0.0001

Data Index: 908612 (C4 (Web Text))

Max Activating Token Index: 768

Text #4

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1463. Min Act: -0.0001

Data Index: 1120454 (C4 (Web Text))

Max Activating Token Index: 371

Text #5

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1354. Min Act: -0.0001

Data Index: 723293 (C4 (Web Text))

Max Activating Token Index: 289

Text #6

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1498. Min Act: -0.0001

Data Index: 65795 (C4 (Web Text))

Max Activating Token Index: 432

Text #7

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1280. Min Act: -0.0001

Data Index: 786921 (C4 (Web Text))

Max Activating Token Index: 860

Text #8

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1269. Min Act: -0.0001

Data Index: 641738 (C4 (Web Text))

Max Activating Token Index: 655

Text #9

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1352. Min Act: -0.0001

Data Index: 207606 (C4 (Web Text))

Max Activating Token Index: 136

Text #10

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1290. Min Act: -0.0001

Data Index: 981480 (C4 (Web Text))

Max Activating Token Index: 61

Text #11

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1379. Min Act: -0.0001

Data Index: 902366 (C4 (Web Text))

Max Activating Token Index: 780

Text #12

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1228. Min Act: -0.0001

Data Index: 1321887 (C4 (Web Text))

Max Activating Token Index: 547

Text #13

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1217. Min Act: -0.0001

Data Index: 287750 (C4 (Web Text))

Max Activating Token Index: 401

Text #14

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1277. Min Act: -0.0001

Data Index: 676757 (C4 (Web Text))

Max Activating Token Index: 121

Text #15

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1313. Min Act: -0.0001

Data Index: 736741 (C4 (Web Text))

Max Activating Token Index: 7

Text #16

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1288. Min Act: -0.0001

Data Index: 984877 (C4 (Web Text))

Max Activating Token Index: 419

Text #17

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1195. Min Act: -0.0001

Data Index: 1620149 (Python Code)

Max Activating Token Index: 5

Text #18

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1122. Min Act: -0.0001

Data Index: 1152676 (C4 (Web Text))

Max Activating Token Index: 709

Text #19

Max Range: 0.1883. Min Range: -0.1883

Max Act: 0.1034. Min Act: -0.0001

Data Index: 1060071 (C4 (Web Text))

Max Activating Token Index: 898