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

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

Neuron 399 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: 0.5645. Min Range: -0.5645

Max Act: 0.5645. Min Act: -0.0000

Data Index: 544904 (C4 (Web Text))

Max Activating Token Index: 842

Text #1

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.4035. Min Act: -0.0000

Data Index: 1146760 (C4 (Web Text))

Max Activating Token Index: 557

Text #2

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3883. Min Act: -0.0000

Data Index: 1416346 (Python Code)

Max Activating Token Index: 102

Text #3

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3756. Min Act: -0.0000

Data Index: 478903 (C4 (Web Text))

Max Activating Token Index: 26

Text #4

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3539. Min Act: -0.0000

Data Index: 637010 (C4 (Web Text))

Max Activating Token Index: 768

Text #5

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3632. Min Act: -0.0000

Data Index: 619826 (C4 (Web Text))

Max Activating Token Index: 962

Text #6

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3699. Min Act: -0.0000

Data Index: 957332 (C4 (Web Text))

Max Activating Token Index: 91

Text #7

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3557. Min Act: -0.0000

Data Index: 719480 (C4 (Web Text))

Max Activating Token Index: 523

Text #8

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3195. Min Act: -0.0000

Data Index: 1281918 (C4 (Web Text))

Max Activating Token Index: 159

Text #9

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3392. Min Act: -0.0000

Data Index: 17543 (C4 (Web Text))

Max Activating Token Index: 41

Text #10

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3416. Min Act: -0.0000

Data Index: 950391 (C4 (Web Text))

Max Activating Token Index: 833

Text #11

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3167. Min Act: -0.0000

Data Index: 1612615 (Python Code)

Max Activating Token Index: 113

Text #12

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3022. Min Act: -0.0000

Data Index: 1559651 (Python Code)

Max Activating Token Index: 879

Text #13

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3077. Min Act: -0.0000

Data Index: 1075098 (C4 (Web Text))

Max Activating Token Index: 816

Text #14

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.2863. Min Act: -0.0000

Data Index: 529878 (C4 (Web Text))

Max Activating Token Index: 718

Text #15

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.2901. Min Act: -0.0000

Data Index: 1303516 (C4 (Web Text))

Max Activating Token Index: 359

Text #16

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.2962. Min Act: -0.0000

Data Index: 207624 (C4 (Web Text))

Max Activating Token Index: 598

Text #17

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.2964. Min Act: -0.0000

Data Index: 677768 (C4 (Web Text))

Max Activating Token Index: 559

Text #18

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.3103. Min Act: -0.0000

Data Index: 354401 (C4 (Web Text))

Max Activating Token Index: 1009

Text #19

Max Range: 0.5645. Min Range: -0.5645

Max Act: 0.2778. Min Act: -0.0000

Data Index: 944048 (C4 (Web Text))

Max Activating Token Index: 32