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

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

Neuron 20 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7179. Min Range: -2.7179

Max Act: 2.5875. Min Act: -0.0000

Data Index: 749540 (C4 (Web Text))

Max Activating Token Index: 154

Text #1

Max Range: 2.7179. Min Range: -2.7179

Max Act: 2.7179. Min Act: -0.0000

Data Index: 370675 (C4 (Web Text))

Max Activating Token Index: 841

Text #2

Max Range: 2.7179. Min Range: -2.7179

Max Act: 2.4188. Min Act: -0.0000

Data Index: 1110497 (C4 (Web Text))

Max Activating Token Index: 845

Text #3

Max Range: 2.7179. Min Range: -2.7179

Max Act: 2.4003. Min Act: -0.0000

Data Index: 875113 (C4 (Web Text))

Max Activating Token Index: 304

Text #4

Max Range: 2.7179. Min Range: -2.7179

Max Act: 2.4364. Min Act: -0.0000

Data Index: 1242604 (C4 (Web Text))

Max Activating Token Index: 201

Text #5

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.6767. Min Act: -0.0000

Data Index: 1089065 (C4 (Web Text))

Max Activating Token Index: 268

Text #6

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.8036. Min Act: -0.0000

Data Index: 813260 (C4 (Web Text))

Max Activating Token Index: 951

Text #7

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.5026. Min Act: -0.0000

Data Index: 756541 (C4 (Web Text))

Max Activating Token Index: 284

Text #8

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.6682. Min Act: -0.0000

Data Index: 1542761 (Python Code)

Max Activating Token Index: 247

Text #9

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.6472. Min Act: -0.0000

Data Index: 1536659 (Python Code)

Max Activating Token Index: 378

Text #10

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.5412. Min Act: -0.0000

Data Index: 1397330 (Python Code)

Max Activating Token Index: 758

Text #11

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.5484. Min Act: -0.0000

Data Index: 1468935 (Python Code)

Max Activating Token Index: 989

Text #12

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.4928. Min Act: -0.0000

Data Index: 1009544 (C4 (Web Text))

Max Activating Token Index: 484

Text #13

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.5631. Min Act: -0.0000

Data Index: 1458419 (Python Code)

Max Activating Token Index: 1005

Text #14

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.4633. Min Act: -0.0000

Data Index: 1505386 (Python Code)

Max Activating Token Index: 215

Text #15

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.4233. Min Act: -0.0000

Data Index: 737854 (C4 (Web Text))

Max Activating Token Index: 599

Text #16

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.5442. Min Act: -0.0000

Data Index: 240035 (C4 (Web Text))

Max Activating Token Index: 478

Text #17

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.4429. Min Act: -0.0000

Data Index: 1210573 (C4 (Web Text))

Max Activating Token Index: 78

Text #18

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.5157. Min Act: -0.0000

Data Index: 1459629 (Python Code)

Max Activating Token Index: 584

Text #19

Max Range: 2.7179. Min Range: -2.7179

Max Act: 1.4331. Min Act: -0.0000

Data Index: 131590 (C4 (Web Text))

Max Activating Token Index: 337