Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3768 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.7960. Min Act: -0.0000

Data Index: 1198325 (C4 (Web Text))

Max Activating Token Index: 909

Text #1

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.7616. Min Act: -0.0000

Data Index: 82116 (C4 (Web Text))

Max Activating Token Index: 1008

Text #2

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.7228. Min Act: -0.0000

Data Index: 1038001 (C4 (Web Text))

Max Activating Token Index: 880

Text #3

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.7102. Min Act: -0.0000

Data Index: 527890 (C4 (Web Text))

Max Activating Token Index: 318

Text #4

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.7288. Min Act: -0.0000

Data Index: 952072 (C4 (Web Text))

Max Activating Token Index: 13

Text #5

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.6732. Min Act: -0.0000

Data Index: 522013 (C4 (Web Text))

Max Activating Token Index: 314

Text #6

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.6306. Min Act: -0.0000

Data Index: 386977 (C4 (Web Text))

Max Activating Token Index: 975

Text #7

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.6180. Min Act: -0.0000

Data Index: 1321878 (C4 (Web Text))

Max Activating Token Index: 274

Text #8

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.5780. Min Act: -0.0000

Data Index: 83182 (C4 (Web Text))

Max Activating Token Index: 332

Text #9

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.5814. Min Act: -0.0000

Data Index: 186886 (C4 (Web Text))

Max Activating Token Index: 872

Text #10

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.6203. Min Act: -0.0000

Data Index: 230472 (C4 (Web Text))

Max Activating Token Index: 100

Text #11

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.6029. Min Act: -0.0000

Data Index: 183037 (C4 (Web Text))

Max Activating Token Index: 577

Text #12

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.5181. Min Act: -0.0000

Data Index: 1227717 (C4 (Web Text))

Max Activating Token Index: 195

Text #13

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.5437. Min Act: -0.0000

Data Index: 1249874 (C4 (Web Text))

Max Activating Token Index: 879

Text #14

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.4916. Min Act: -0.0000

Data Index: 1036250 (C4 (Web Text))

Max Activating Token Index: 100

Text #15

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.5080. Min Act: -0.0000

Data Index: 1119973 (C4 (Web Text))

Max Activating Token Index: 203

Text #16

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.5088. Min Act: -0.0000

Data Index: 1036601 (C4 (Web Text))

Max Activating Token Index: 184

Text #17

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.5205. Min Act: -0.0000

Data Index: 492455 (C4 (Web Text))

Max Activating Token Index: 911

Text #18

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.5194. Min Act: -0.0000

Data Index: 873017 (C4 (Web Text))

Max Activating Token Index: 125

Text #19

Max Range: 0.7960. Min Range: -0.7960

Max Act: 0.5088. Min Act: -0.0000

Data Index: 1704 (C4 (Web Text))

Max Activating Token Index: 589