Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1901 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.1034. Min Act: -0.0002

Data Index: 438412 (C4 (Web Text))

Max Activating Token Index: 38

Text #1

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.1018. Min Act: -0.0002

Data Index: 1245874 (C4 (Web Text))

Max Activating Token Index: 514

Text #2

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0957. Min Act: -0.0002

Data Index: 229109 (C4 (Web Text))

Max Activating Token Index: 259

Text #3

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0930. Min Act: -0.0002

Data Index: 1581260 (Python Code)

Max Activating Token Index: 371

Text #4

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0920. Min Act: -0.0002

Data Index: 94708 (C4 (Web Text))

Max Activating Token Index: 165

Text #5

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0839. Min Act: -0.0002

Data Index: 1556453 (Python Code)

Max Activating Token Index: 188

Text #6

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0842. Min Act: -0.0002

Data Index: 1571183 (Python Code)

Max Activating Token Index: 241

Text #7

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0840. Min Act: -0.0001

Data Index: 634285 (C4 (Web Text))

Max Activating Token Index: 50

Text #8

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0801. Min Act: -0.0002

Data Index: 907401 (C4 (Web Text))

Max Activating Token Index: 243

Text #9

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0847. Min Act: -0.0002

Data Index: 1047378 (C4 (Web Text))

Max Activating Token Index: 787

Text #10

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0815. Min Act: -0.0001

Data Index: 1585820 (Python Code)

Max Activating Token Index: 540

Text #11

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0808. Min Act: -0.0002

Data Index: 392368 (C4 (Web Text))

Max Activating Token Index: 174

Text #12

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0799. Min Act: -0.0002

Data Index: 161524 (C4 (Web Text))

Max Activating Token Index: 78

Text #13

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0819. Min Act: -0.0001

Data Index: 1521000 (Python Code)

Max Activating Token Index: 100

Text #14

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0793. Min Act: -0.0002

Data Index: 1655400 (Python Code)

Max Activating Token Index: 158

Text #15

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0803. Min Act: -0.0002

Data Index: 1571819 (Python Code)

Max Activating Token Index: 967

Text #16

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0786. Min Act: -0.0002

Data Index: 1484290 (Python Code)

Max Activating Token Index: 724

Text #17

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0793. Min Act: -0.0002

Data Index: 1549045 (Python Code)

Max Activating Token Index: 306

Text #18

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0740. Min Act: -0.0002

Data Index: 963473 (C4 (Web Text))

Max Activating Token Index: 52

Text #19

Max Range: 0.1034. Min Range: -0.1034

Max Act: 0.0754. Min Act: -0.0002

Data Index: 1040029 (C4 (Web Text))

Max Activating Token Index: 59