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

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

Neuron 436 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.4818. Min Act: -0.0000

Data Index: 1611465 (Python Code)

Max Activating Token Index: 597

Text #1

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.4666. Min Act: -0.0000

Data Index: 249181 (C4 (Web Text))

Max Activating Token Index: 561

Text #2

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.4047. Min Act: -0.0000

Data Index: 1374422 (Python Code)

Max Activating Token Index: 480

Text #3

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.2912. Min Act: -0.0000

Data Index: 670589 (C4 (Web Text))

Max Activating Token Index: 865

Text #4

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.2445. Min Act: -0.0000

Data Index: 1648324 (Python Code)

Max Activating Token Index: 489

Text #5

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.2487. Min Act: -0.0000

Data Index: 435285 (C4 (Web Text))

Max Activating Token Index: 212

Text #6

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.1663. Min Act: -0.0000

Data Index: 305879 (C4 (Web Text))

Max Activating Token Index: 891

Text #7

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.1027. Min Act: -0.0001

Data Index: 1543886 (Python Code)

Max Activating Token Index: 688

Text #8

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.1330. Min Act: -0.0000

Data Index: 1389998 (Python Code)

Max Activating Token Index: 221

Text #9

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.1197. Min Act: -0.0000

Data Index: 1604548 (Python Code)

Max Activating Token Index: 587

Text #10

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.0423. Min Act: -0.0000

Data Index: 763364 (C4 (Web Text))

Max Activating Token Index: 196

Text #11

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.0620. Min Act: -0.0000

Data Index: 15468 (C4 (Web Text))

Max Activating Token Index: 112

Text #12

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.0957. Min Act: -0.0000

Data Index: 358037 (C4 (Web Text))

Max Activating Token Index: 730

Text #13

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.1124. Min Act: -0.0000

Data Index: 1349068 (C4 (Web Text))

Max Activating Token Index: 111

Text #14

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.0548. Min Act: -0.0000

Data Index: 815245 (C4 (Web Text))

Max Activating Token Index: 862

Text #15

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.0476. Min Act: -0.0000

Data Index: 34915 (C4 (Web Text))

Max Activating Token Index: 575

Text #16

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.0045. Min Act: -0.0000

Data Index: 1495453 (Python Code)

Max Activating Token Index: 72

Text #17

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.0342. Min Act: -0.0000

Data Index: 247457 (C4 (Web Text))

Max Activating Token Index: 831

Text #18

Max Range: 1.4818. Min Range: -1.4818

Max Act: 0.9631. Min Act: -0.0000

Data Index: 28987 (C4 (Web Text))

Max Activating Token Index: 399

Text #19

Max Range: 1.4818. Min Range: -1.4818

Max Act: 1.0442. Min Act: -0.0000

Data Index: 165038 (C4 (Web Text))

Max Activating Token Index: 870