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

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

Neuron 2061 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: 1.4737. Min Range: -1.4737

Max Act: 1.4678. Min Act: -0.0001

Data Index: 886158 (C4 (Web Text))

Max Activating Token Index: 407

Text #1

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.4737. Min Act: -0.0001

Data Index: 163654 (C4 (Web Text))

Max Activating Token Index: 795

Text #2

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.3862. Min Act: -0.0001

Data Index: 701394 (C4 (Web Text))

Max Activating Token Index: 460

Text #3

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.4223. Min Act: -0.0001

Data Index: 919904 (C4 (Web Text))

Max Activating Token Index: 803

Text #4

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.2699. Min Act: -0.0001

Data Index: 193887 (C4 (Web Text))

Max Activating Token Index: 802

Text #5

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.2500. Min Act: -0.0001

Data Index: 1485363 (Python Code)

Max Activating Token Index: 715

Text #6

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.4162. Min Act: -0.0001

Data Index: 813870 (C4 (Web Text))

Max Activating Token Index: 410

Text #7

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.2029. Min Act: -0.0001

Data Index: 1492345 (Python Code)

Max Activating Token Index: 811

Text #8

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.3743. Min Act: -0.0001

Data Index: 1461134 (Python Code)

Max Activating Token Index: 891

Text #9

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.2604. Min Act: -0.0001

Data Index: 901761 (C4 (Web Text))

Max Activating Token Index: 15

Text #10

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.3298. Min Act: -0.0001

Data Index: 1361068 (Python Code)

Max Activating Token Index: 396

Text #11

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.3092. Min Act: -0.0001

Data Index: 1203204 (C4 (Web Text))

Max Activating Token Index: 446

Text #12

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.4165. Min Act: -0.0001

Data Index: 1067172 (C4 (Web Text))

Max Activating Token Index: 903

Text #13

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.2460. Min Act: -0.0001

Data Index: 139906 (C4 (Web Text))

Max Activating Token Index: 935

Text #14

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.2343. Min Act: -0.0001

Data Index: 1132151 (C4 (Web Text))

Max Activating Token Index: 42

Text #15

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.3138. Min Act: -0.0001

Data Index: 160804 (C4 (Web Text))

Max Activating Token Index: 1010

Text #16

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.1765. Min Act: -0.0001

Data Index: 392179 (C4 (Web Text))

Max Activating Token Index: 756

Text #17

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.1549. Min Act: -0.0001

Data Index: 1458937 (Python Code)

Max Activating Token Index: 593

Text #18

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.1860. Min Act: -0.0001

Data Index: 354408 (C4 (Web Text))

Max Activating Token Index: 150

Text #19

Max Range: 1.4737. Min Range: -1.4737

Max Act: 1.2138. Min Act: -0.0001

Data Index: 1446492 (Python Code)

Max Activating Token Index: 476