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

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

Neuron 2019 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7250. Min Range: -3.7250

Max Act: 3.7250. Min Act: -0.0000

Data Index: 432525 (C4 (Web Text))

Max Activating Token Index: 506

Text #1

Max Range: 3.7250. Min Range: -3.7250

Max Act: 3.0272. Min Act: -0.0000

Data Index: 408483 (C4 (Web Text))

Max Activating Token Index: 777

Text #2

Max Range: 3.7250. Min Range: -3.7250

Max Act: 2.1718. Min Act: -0.0000

Data Index: 95370 (C4 (Web Text))

Max Activating Token Index: 896

Text #3

Max Range: 3.7250. Min Range: -3.7250

Max Act: 2.3152. Min Act: -0.0000

Data Index: 884560 (C4 (Web Text))

Max Activating Token Index: 294

Text #4

Max Range: 3.7250. Min Range: -3.7250

Max Act: 2.0568. Min Act: -0.0000

Data Index: 249199 (C4 (Web Text))

Max Activating Token Index: 866

Text #5

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.9293. Min Act: -0.0000

Data Index: 1113112 (C4 (Web Text))

Max Activating Token Index: 157

Text #6

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.7998. Min Act: -0.0000

Data Index: 826098 (C4 (Web Text))

Max Activating Token Index: 745

Text #7

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.6143. Min Act: -0.0000

Data Index: 440828 (C4 (Web Text))

Max Activating Token Index: 30

Text #8

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.7632. Min Act: -0.0000

Data Index: 792839 (C4 (Web Text))

Max Activating Token Index: 917

Text #9

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.5452. Min Act: -0.0000

Data Index: 821195 (C4 (Web Text))

Max Activating Token Index: 706

Text #10

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.2969. Min Act: -0.0000

Data Index: 973871 (C4 (Web Text))

Max Activating Token Index: 589

Text #11

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.2123. Min Act: -0.0000

Data Index: 109729 (C4 (Web Text))

Max Activating Token Index: 151

Text #12

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.1612. Min Act: 0.0000

Data Index: 811258 (C4 (Web Text))

Max Activating Token Index: 637

Text #13

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.0937. Min Act: -0.0000

Data Index: 841956 (C4 (Web Text))

Max Activating Token Index: 216

Text #14

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.2315. Min Act: -0.0000

Data Index: 68687 (C4 (Web Text))

Max Activating Token Index: 145

Text #15

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.0864. Min Act: -0.0000

Data Index: 802180 (C4 (Web Text))

Max Activating Token Index: 908

Text #16

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.1005. Min Act: -0.0000

Data Index: 512308 (C4 (Web Text))

Max Activating Token Index: 505

Text #17

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.1284. Min Act: -0.0000

Data Index: 500788 (C4 (Web Text))

Max Activating Token Index: 218

Text #18

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.1195. Min Act: -0.0000

Data Index: 1209026 (C4 (Web Text))

Max Activating Token Index: 148

Text #19

Max Range: 3.7250. Min Range: -3.7250

Max Act: 1.1895. Min Act: -0.0000

Data Index: 1549807 (Python Code)

Max Activating Token Index: 818