Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3521 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.1280. Min Range: -1.1280

Max Act: 1.1280. Min Act: -0.0001

Data Index: 520568 (C4 (Web Text))

Max Activating Token Index: 782

Text #1

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8999. Min Act: -0.0001

Data Index: 956646 (C4 (Web Text))

Max Activating Token Index: 523

Text #2

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.9749. Min Act: -0.0001

Data Index: 110921 (C4 (Web Text))

Max Activating Token Index: 214

Text #3

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8947. Min Act: -0.0001

Data Index: 103994 (C4 (Web Text))

Max Activating Token Index: 996

Text #4

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8704. Min Act: -0.0001

Data Index: 1010316 (C4 (Web Text))

Max Activating Token Index: 330

Text #5

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8450. Min Act: -0.0001

Data Index: 1258062 (C4 (Web Text))

Max Activating Token Index: 773

Text #6

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8250. Min Act: -0.0001

Data Index: 109322 (C4 (Web Text))

Max Activating Token Index: 46

Text #7

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8471. Min Act: -0.0001

Data Index: 1442075 (Python Code)

Max Activating Token Index: 930

Text #8

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8365. Min Act: -0.0001

Data Index: 145293 (C4 (Web Text))

Max Activating Token Index: 713

Text #9

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8307. Min Act: -0.0001

Data Index: 1127804 (C4 (Web Text))

Max Activating Token Index: 424

Text #10

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8549. Min Act: -0.0001

Data Index: 382491 (C4 (Web Text))

Max Activating Token Index: 601

Text #11

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8577. Min Act: -0.0001

Data Index: 289611 (C4 (Web Text))

Max Activating Token Index: 323

Text #12

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8518. Min Act: -0.0001

Data Index: 1373438 (Python Code)

Max Activating Token Index: 456

Text #13

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.7646. Min Act: -0.0001

Data Index: 258058 (C4 (Web Text))

Max Activating Token Index: 411

Text #14

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8444. Min Act: -0.0001

Data Index: 972742 (C4 (Web Text))

Max Activating Token Index: 304

Text #15

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.7409. Min Act: -0.0001

Data Index: 489870 (C4 (Web Text))

Max Activating Token Index: 57

Text #16

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.8129. Min Act: -0.0001

Data Index: 1570170 (Python Code)

Max Activating Token Index: 569

Text #17

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.7425. Min Act: -0.0001

Data Index: 1080798 (C4 (Web Text))

Max Activating Token Index: 239

Text #18

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.7370. Min Act: -0.0001

Data Index: 1433183 (Python Code)

Max Activating Token Index: 139

Text #19

Max Range: 1.1280. Min Range: -1.1280

Max Act: 0.7261. Min Act: -0.0001

Data Index: 948711 (C4 (Web Text))

Max Activating Token Index: 240