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

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

Neuron 4815 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.7523. Min Act: -0.0000

Data Index: 1077426 (C4 (Web Text))

Max Activating Token Index: 1022

Text #1

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.7368. Min Act: -0.0000

Data Index: 130843 (C4 (Web Text))

Max Activating Token Index: 519

Text #2

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.7058. Min Act: -0.0000

Data Index: 1359129 (C4 (Web Text))

Max Activating Token Index: 85

Text #3

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5911. Min Act: -0.0000

Data Index: 709810 (C4 (Web Text))

Max Activating Token Index: 231

Text #4

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5826. Min Act: -0.0000

Data Index: 1441082 (Python Code)

Max Activating Token Index: 741

Text #5

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5963. Min Act: -0.0000

Data Index: 1360914 (Python Code)

Max Activating Token Index: 279

Text #6

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5812. Min Act: 0.0000

Data Index: 745999 (C4 (Web Text))

Max Activating Token Index: 249

Text #7

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.6048. Min Act: -0.0000

Data Index: 261709 (C4 (Web Text))

Max Activating Token Index: 189

Text #8

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5635. Min Act: -0.0000

Data Index: 1336654 (C4 (Web Text))

Max Activating Token Index: 316

Text #9

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5560. Min Act: -0.0000

Data Index: 1400487 (Python Code)

Max Activating Token Index: 110

Text #10

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5640. Min Act: -0.0000

Data Index: 538415 (C4 (Web Text))

Max Activating Token Index: 347

Text #11

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5412. Min Act: -0.0000

Data Index: 1147325 (C4 (Web Text))

Max Activating Token Index: 418

Text #12

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5327. Min Act: -0.0000

Data Index: 18468 (C4 (Web Text))

Max Activating Token Index: 768

Text #13

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5286. Min Act: -0.0000

Data Index: 149095 (C4 (Web Text))

Max Activating Token Index: 567

Text #14

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5212. Min Act: -0.0000

Data Index: 147726 (C4 (Web Text))

Max Activating Token Index: 66

Text #15

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5440. Min Act: -0.0000

Data Index: 1579424 (Python Code)

Max Activating Token Index: 48

Text #16

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5048. Min Act: -0.0000

Data Index: 805590 (C4 (Web Text))

Max Activating Token Index: 166

Text #17

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5169. Min Act: -0.0000

Data Index: 1404145 (Python Code)

Max Activating Token Index: 67

Text #18

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5127. Min Act: -0.0000

Data Index: 1193693 (C4 (Web Text))

Max Activating Token Index: 150

Text #19

Max Range: 0.7523. Min Range: -0.7523

Max Act: 0.5092. Min Act: -0.0000

Data Index: 847622 (C4 (Web Text))

Max Activating Token Index: 12