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

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

Neuron 4100 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.6573. Min Act: -0.0000

Data Index: 1608074 (Python Code)

Max Activating Token Index: 120

Text #1

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.5021. Min Act: -0.0000

Data Index: 1495393 (Python Code)

Max Activating Token Index: 755

Text #2

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.4353. Min Act: -0.0000

Data Index: 690881 (C4 (Web Text))

Max Activating Token Index: 752

Text #3

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.4855. Min Act: -0.0000

Data Index: 1645319 (Python Code)

Max Activating Token Index: 822

Text #4

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.4182. Min Act: -0.0000

Data Index: 119731 (C4 (Web Text))

Max Activating Token Index: 410

Text #5

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.4460. Min Act: -0.0000

Data Index: 1582564 (Python Code)

Max Activating Token Index: 906

Text #6

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.4299. Min Act: -0.0000

Data Index: 1619552 (Python Code)

Max Activating Token Index: 1003

Text #7

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.4161. Min Act: -0.0000

Data Index: 1375620 (Python Code)

Max Activating Token Index: 425

Text #8

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.4095. Min Act: -0.0000

Data Index: 715499 (C4 (Web Text))

Max Activating Token Index: 439

Text #9

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.4177. Min Act: -0.0000

Data Index: 1593119 (Python Code)

Max Activating Token Index: 102

Text #10

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.4149. Min Act: -0.0000

Data Index: 1634700 (Python Code)

Max Activating Token Index: 252

Text #11

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.3879. Min Act: -0.0000

Data Index: 1386286 (Python Code)

Max Activating Token Index: 780

Text #12

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.3907. Min Act: -0.0000

Data Index: 1370053 (Python Code)

Max Activating Token Index: 101

Text #13

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.3504. Min Act: -0.0000

Data Index: 1578046 (Python Code)

Max Activating Token Index: 495

Text #14

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.3726. Min Act: -0.0000

Data Index: 1480580 (Python Code)

Max Activating Token Index: 96

Text #15

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.3254. Min Act: -0.0000

Data Index: 1457386 (Python Code)

Max Activating Token Index: 417

Text #16

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.3433. Min Act: -0.0000

Data Index: 729265 (C4 (Web Text))

Max Activating Token Index: 252

Text #17

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.3579. Min Act: -0.0000

Data Index: 1387926 (Python Code)

Max Activating Token Index: 985

Text #18

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.3452. Min Act: -0.0000

Data Index: 62913 (C4 (Web Text))

Max Activating Token Index: 826

Text #19

Max Range: 0.6573. Min Range: -0.6573

Max Act: 0.3295. Min Act: 0.0001

Data Index: 1431404 (Python Code)

Max Activating Token Index: 223