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

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

Neuron 4777 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.9998. Min Range: -5.9998

Max Act: 5.9998. Min Act: -0.0000

Data Index: 202372 (C4 (Web Text))

Max Activating Token Index: 777

Text #1

Max Range: 5.9998. Min Range: -5.9998

Max Act: 5.7640. Min Act: -0.0000

Data Index: 1073089 (C4 (Web Text))

Max Activating Token Index: 330

Text #2

Max Range: 5.9998. Min Range: -5.9998

Max Act: 5.3140. Min Act: -0.0000

Data Index: 1241377 (C4 (Web Text))

Max Activating Token Index: 269

Text #3

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.9192. Min Act: -0.0000

Data Index: 1642537 (Python Code)

Max Activating Token Index: 969

Text #4

Max Range: 5.9998. Min Range: -5.9998

Max Act: 5.1135. Min Act: -0.0000

Data Index: 895162 (C4 (Web Text))

Max Activating Token Index: 220

Text #5

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.8334. Min Act: -0.0000

Data Index: 1525655 (Python Code)

Max Activating Token Index: 246

Text #6

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.7576. Min Act: -0.0000

Data Index: 131126 (C4 (Web Text))

Max Activating Token Index: 439

Text #7

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.6462. Min Act: -0.0000

Data Index: 1171239 (C4 (Web Text))

Max Activating Token Index: 307

Text #8

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.4302. Min Act: -0.0000

Data Index: 1422728 (Python Code)

Max Activating Token Index: 102

Text #9

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.5283. Min Act: -0.0000

Data Index: 995485 (C4 (Web Text))

Max Activating Token Index: 1021

Text #10

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.4953. Min Act: -0.0000

Data Index: 1351620 (C4 (Web Text))

Max Activating Token Index: 542

Text #11

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.4882. Min Act: -0.0000

Data Index: 1066942 (C4 (Web Text))

Max Activating Token Index: 496

Text #12

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.4824. Min Act: -0.0000

Data Index: 1068773 (C4 (Web Text))

Max Activating Token Index: 232

Text #13

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.4634. Min Act: -0.0000

Data Index: 1086756 (C4 (Web Text))

Max Activating Token Index: 1019

Text #14

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.6501. Min Act: -0.0000

Data Index: 152620 (C4 (Web Text))

Max Activating Token Index: 194

Text #15

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.6328. Min Act: -0.0000

Data Index: 419980 (C4 (Web Text))

Max Activating Token Index: 813

Text #16

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.2432. Min Act: -0.0000

Data Index: 283831 (C4 (Web Text))

Max Activating Token Index: 540

Text #17

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.2380. Min Act: -0.0000

Data Index: 106636 (C4 (Web Text))

Max Activating Token Index: 139

Text #18

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.3630. Min Act: -0.0000

Data Index: 722040 (C4 (Web Text))

Max Activating Token Index: 31

Text #19

Max Range: 5.9998. Min Range: -5.9998

Max Act: 4.3564. Min Act: -0.0000

Data Index: 1366004 (Python Code)

Max Activating Token Index: 922