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

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

Neuron 5050 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2726. Min Act: -0.0001

Data Index: 104017 (C4 (Web Text))

Max Activating Token Index: 232

Text #1

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2771. Min Act: -0.0001

Data Index: 1080028 (C4 (Web Text))

Max Activating Token Index: 556

Text #2

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2703. Min Act: -0.0001

Data Index: 879176 (C4 (Web Text))

Max Activating Token Index: 134

Text #3

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2525. Min Act: -0.0001

Data Index: 1225204 (C4 (Web Text))

Max Activating Token Index: 740

Text #4

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2164. Min Act: -0.0001

Data Index: 1641388 (Python Code)

Max Activating Token Index: 218

Text #5

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2257. Min Act: -0.0001

Data Index: 370713 (C4 (Web Text))

Max Activating Token Index: 215

Text #6

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2299. Min Act: -0.0001

Data Index: 287986 (C4 (Web Text))

Max Activating Token Index: 34

Text #7

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2382. Min Act: -0.0001

Data Index: 1228043 (C4 (Web Text))

Max Activating Token Index: 23

Text #8

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2157. Min Act: -0.0001

Data Index: 178431 (C4 (Web Text))

Max Activating Token Index: 190

Text #9

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2366. Min Act: -0.0001

Data Index: 128518 (C4 (Web Text))

Max Activating Token Index: 791

Text #10

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.1948. Min Act: -0.0001

Data Index: 1597311 (Python Code)

Max Activating Token Index: 96

Text #11

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2092. Min Act: -0.0001

Data Index: 27872 (C4 (Web Text))

Max Activating Token Index: 478

Text #12

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2007. Min Act: -0.0001

Data Index: 1642474 (Python Code)

Max Activating Token Index: 1003

Text #13

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2025. Min Act: -0.0001

Data Index: 737232 (C4 (Web Text))

Max Activating Token Index: 1013

Text #14

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.1974. Min Act: -0.0001

Data Index: 1197664 (C4 (Web Text))

Max Activating Token Index: 658

Text #15

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2055. Min Act: -0.0001

Data Index: 1505801 (Python Code)

Max Activating Token Index: 992

Text #16

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2089. Min Act: -0.0001

Data Index: 1294488 (C4 (Web Text))

Max Activating Token Index: 648

Text #17

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.2052. Min Act: -0.0001

Data Index: 1104270 (C4 (Web Text))

Max Activating Token Index: 986

Text #18

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.1980. Min Act: -0.0001

Data Index: 310343 (C4 (Web Text))

Max Activating Token Index: 360

Text #19

Max Range: 0.2771. Min Range: -0.2771

Max Act: 0.1950. Min Act: -0.0001

Data Index: 555290 (C4 (Web Text))

Max Activating Token Index: 5