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

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

Neuron 2105 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.1058. Min Range: -2.1058

Max Act: 2.1058. Min Act: -0.0000

Data Index: 1591355 (Python Code)

Max Activating Token Index: 278

Text #1

Max Range: 2.1058. Min Range: -2.1058

Max Act: 2.0735. Min Act: -0.0000

Data Index: 1195087 (C4 (Web Text))

Max Activating Token Index: 683

Text #2

Max Range: 2.1058. Min Range: -2.1058

Max Act: 2.0260. Min Act: -0.0000

Data Index: 707042 (C4 (Web Text))

Max Activating Token Index: 74

Text #3

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.9493. Min Act: -0.0000

Data Index: 165975 (C4 (Web Text))

Max Activating Token Index: 759

Text #4

Max Range: 2.1058. Min Range: -2.1058

Max Act: 2.0615. Min Act: -0.0000

Data Index: 648560 (C4 (Web Text))

Max Activating Token Index: 853

Text #5

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.9289. Min Act: -0.0000

Data Index: 1487595 (Python Code)

Max Activating Token Index: 690

Text #6

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.9224. Min Act: -0.0000

Data Index: 559089 (C4 (Web Text))

Max Activating Token Index: 398

Text #7

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.9010. Min Act: -0.0000

Data Index: 1466945 (Python Code)

Max Activating Token Index: 167

Text #8

Max Range: 2.1058. Min Range: -2.1058

Max Act: 2.0058. Min Act: -0.0000

Data Index: 1468181 (Python Code)

Max Activating Token Index: 790

Text #9

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.7688. Min Act: -0.0000

Data Index: 1634140 (Python Code)

Max Activating Token Index: 209

Text #10

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.8527. Min Act: -0.0000

Data Index: 219910 (C4 (Web Text))

Max Activating Token Index: 434

Text #11

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.7396. Min Act: -0.0000

Data Index: 14922 (C4 (Web Text))

Max Activating Token Index: 667

Text #12

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.8323. Min Act: -0.0000

Data Index: 1308014 (C4 (Web Text))

Max Activating Token Index: 100

Text #13

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.9285. Min Act: -0.0000

Data Index: 1561388 (Python Code)

Max Activating Token Index: 591

Text #14

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.8162. Min Act: -0.0000

Data Index: 1390210 (Python Code)

Max Activating Token Index: 423

Text #15

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.8076. Min Act: -0.0000

Data Index: 1483561 (Python Code)

Max Activating Token Index: 524

Text #16

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.9182. Min Act: -0.0000

Data Index: 1543426 (Python Code)

Max Activating Token Index: 773

Text #17

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.9072. Min Act: -0.0000

Data Index: 1458216 (Python Code)

Max Activating Token Index: 160

Text #18

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.7033. Min Act: -0.0000

Data Index: 1363798 (Python Code)

Max Activating Token Index: 804

Text #19

Max Range: 2.1058. Min Range: -2.1058

Max Act: 1.7881. Min Act: -0.0000

Data Index: 1577977 (Python Code)

Max Activating Token Index: 165