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

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

Neuron 4397 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.8871. Min Act: -0.0001

Data Index: 20996 (C4 (Web Text))

Max Activating Token Index: 711

Text #1

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.5611. Min Act: -0.0001

Data Index: 1143247 (C4 (Web Text))

Max Activating Token Index: 700

Text #2

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.3673. Min Act: -0.0001

Data Index: 1607752 (Python Code)

Max Activating Token Index: 99

Text #3

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.3288. Min Act: -0.0001

Data Index: 1194575 (C4 (Web Text))

Max Activating Token Index: 1018

Text #4

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.4195. Min Act: -0.0000

Data Index: 20623 (C4 (Web Text))

Max Activating Token Index: 424

Text #5

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.3300. Min Act: -0.0001

Data Index: 632738 (C4 (Web Text))

Max Activating Token Index: 375

Text #6

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.2075. Min Act: -0.0000

Data Index: 1361295 (Python Code)

Max Activating Token Index: 551

Text #7

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.1975. Min Act: -0.0000

Data Index: 380947 (C4 (Web Text))

Max Activating Token Index: 207

Text #8

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.3332. Min Act: -0.0000

Data Index: 889130 (C4 (Web Text))

Max Activating Token Index: 541

Text #9

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.2201. Min Act: -0.0000

Data Index: 1002733 (C4 (Web Text))

Max Activating Token Index: 339

Text #10

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.2758. Min Act: -0.0001

Data Index: 1164456 (C4 (Web Text))

Max Activating Token Index: 414

Text #11

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.2163. Min Act: -0.0001

Data Index: 644321 (C4 (Web Text))

Max Activating Token Index: 563

Text #12

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.2530. Min Act: -0.0001

Data Index: 1067617 (C4 (Web Text))

Max Activating Token Index: 450

Text #13

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.2449. Min Act: -0.0001

Data Index: 780157 (C4 (Web Text))

Max Activating Token Index: 834

Text #14

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.2401. Min Act: -0.0001

Data Index: 1106705 (C4 (Web Text))

Max Activating Token Index: 405

Text #15

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.2118. Min Act: -0.0000

Data Index: 436758 (C4 (Web Text))

Max Activating Token Index: 105

Text #16

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.1745. Min Act: -0.0001

Data Index: 253712 (C4 (Web Text))

Max Activating Token Index: 8

Text #17

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.1020. Min Act: -0.0001

Data Index: 427271 (C4 (Web Text))

Max Activating Token Index: 602

Text #18

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.0511. Min Act: -0.0000

Data Index: 656752 (C4 (Web Text))

Max Activating Token Index: 844

Text #19

Max Range: 1.8871. Min Range: -1.8871

Max Act: 1.1526. Min Act: -0.0001

Data Index: 955612 (C4 (Web Text))

Max Activating Token Index: 442