Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1700 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.3355. Min Act: -0.0001

Data Index: 1172093 (C4 (Web Text))

Max Activating Token Index: 425

Text #1

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.3135. Min Act: -0.0001

Data Index: 822111 (C4 (Web Text))

Max Activating Token Index: 999

Text #2

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.3018. Min Act: -0.0001

Data Index: 1211997 (C4 (Web Text))

Max Activating Token Index: 143

Text #3

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.3098. Min Act: -0.0001

Data Index: 706820 (C4 (Web Text))

Max Activating Token Index: 870

Text #4

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.3070. Min Act: -0.0001

Data Index: 791516 (C4 (Web Text))

Max Activating Token Index: 357

Text #5

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.3038. Min Act: -0.0001

Data Index: 1567877 (Python Code)

Max Activating Token Index: 568

Text #6

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.3013. Min Act: -0.0001

Data Index: 163818 (C4 (Web Text))

Max Activating Token Index: 17

Text #7

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.2869. Min Act: -0.0001

Data Index: 583009 (C4 (Web Text))

Max Activating Token Index: 911

Text #8

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.3076. Min Act: -0.0001

Data Index: 442041 (C4 (Web Text))

Max Activating Token Index: 590

Text #9

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.2837. Min Act: -0.0001

Data Index: 1052555 (C4 (Web Text))

Max Activating Token Index: 692

Text #10

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.2827. Min Act: -0.0001

Data Index: 1030102 (C4 (Web Text))

Max Activating Token Index: 990

Text #11

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.3138. Min Act: -0.0001

Data Index: 1625591 (Python Code)

Max Activating Token Index: 731

Text #12

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.2646. Min Act: -0.0001

Data Index: 930348 (C4 (Web Text))

Max Activating Token Index: 968

Text #13

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.2638. Min Act: -0.0001

Data Index: 313774 (C4 (Web Text))

Max Activating Token Index: 95

Text #14

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.2530. Min Act: -0.0001

Data Index: 974435 (C4 (Web Text))

Max Activating Token Index: 633

Text #15

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.2728. Min Act: -0.0001

Data Index: 129493 (C4 (Web Text))

Max Activating Token Index: 233

Text #16

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.2603. Min Act: -0.0001

Data Index: 1567624 (Python Code)

Max Activating Token Index: 569

Text #17

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.2772. Min Act: -0.0001

Data Index: 1149585 (C4 (Web Text))

Max Activating Token Index: 984

Text #18

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.2762. Min Act: -0.0001

Data Index: 174596 (C4 (Web Text))

Max Activating Token Index: 718

Text #19

Max Range: 0.3355. Min Range: -0.3355

Max Act: 0.2678. Min Act: -0.0001

Data Index: 94646 (C4 (Web Text))

Max Activating Token Index: 996