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

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

Neuron 3344 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.5944. Min Act: -0.0000

Data Index: 1581966 (Python Code)

Max Activating Token Index: 202

Text #1

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.6223. Min Act: -0.0000

Data Index: 1352413 (C4 (Web Text))

Max Activating Token Index: 919

Text #2

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.5524. Min Act: -0.0000

Data Index: 228665 (C4 (Web Text))

Max Activating Token Index: 914

Text #3

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.5185. Min Act: -0.0000

Data Index: 840391 (C4 (Web Text))

Max Activating Token Index: 699

Text #4

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.4720. Min Act: -0.0000

Data Index: 1599715 (Python Code)

Max Activating Token Index: 122

Text #5

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.5221. Min Act: -0.0000

Data Index: 1484972 (Python Code)

Max Activating Token Index: 224

Text #6

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.4982. Min Act: -0.0000

Data Index: 1634658 (Python Code)

Max Activating Token Index: 804

Text #7

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.5087. Min Act: -0.0000

Data Index: 1429528 (Python Code)

Max Activating Token Index: 415

Text #8

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.4787. Min Act: -0.0000

Data Index: 504074 (C4 (Web Text))

Max Activating Token Index: 511

Text #9

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.5078. Min Act: -0.0000

Data Index: 1403033 (Python Code)

Max Activating Token Index: 101

Text #10

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.4931. Min Act: -0.0000

Data Index: 260565 (C4 (Web Text))

Max Activating Token Index: 465

Text #11

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.5091. Min Act: -0.0000

Data Index: 1434415 (Python Code)

Max Activating Token Index: 128

Text #12

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.5069. Min Act: -0.0000

Data Index: 1624155 (Python Code)

Max Activating Token Index: 904

Text #13

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.5245. Min Act: -0.0000

Data Index: 884479 (C4 (Web Text))

Max Activating Token Index: 779

Text #14

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.5001. Min Act: -0.0000

Data Index: 1455474 (Python Code)

Max Activating Token Index: 720

Text #15

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.5020. Min Act: -0.0000

Data Index: 1651004 (Python Code)

Max Activating Token Index: 435

Text #16

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.4988. Min Act: -0.0000

Data Index: 917768 (C4 (Web Text))

Max Activating Token Index: 549

Text #17

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.5186. Min Act: -0.0000

Data Index: 232128 (C4 (Web Text))

Max Activating Token Index: 838

Text #18

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.4994. Min Act: -0.0000

Data Index: 304003 (C4 (Web Text))

Max Activating Token Index: 671

Text #19

Max Range: 0.6223. Min Range: -0.6223

Max Act: 0.4748. Min Act: -0.0000

Data Index: 547327 (C4 (Web Text))

Max Activating Token Index: 752