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

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

Neuron 2395 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.8181. Min Act: -0.0000

Data Index: 266815 (C4 (Web Text))

Max Activating Token Index: 861

Text #1

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.8073. Min Act: -0.0000

Data Index: 79583 (C4 (Web Text))

Max Activating Token Index: 977

Text #2

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.7821. Min Act: -0.0000

Data Index: 436712 (C4 (Web Text))

Max Activating Token Index: 679

Text #3

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.7965. Min Act: -0.0000

Data Index: 492927 (C4 (Web Text))

Max Activating Token Index: 436

Text #4

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.7524. Min Act: -0.0000

Data Index: 706328 (C4 (Web Text))

Max Activating Token Index: 810

Text #5

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.7464. Min Act: -0.0000

Data Index: 544416 (C4 (Web Text))

Max Activating Token Index: 624

Text #6

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.6894. Min Act: -0.0000

Data Index: 1022958 (C4 (Web Text))

Max Activating Token Index: 119

Text #7

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.6787. Min Act: -0.0000

Data Index: 866502 (C4 (Web Text))

Max Activating Token Index: 349

Text #8

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.7086. Min Act: -0.0000

Data Index: 1071504 (C4 (Web Text))

Max Activating Token Index: 901

Text #9

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.7095. Min Act: -0.0000

Data Index: 1333168 (C4 (Web Text))

Max Activating Token Index: 878

Text #10

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.6999. Min Act: 0.0001

Data Index: 239189 (C4 (Web Text))

Max Activating Token Index: 1009

Text #11

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.6777. Min Act: -0.0000

Data Index: 424946 (C4 (Web Text))

Max Activating Token Index: 560

Text #12

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.6746. Min Act: -0.0000

Data Index: 1111829 (C4 (Web Text))

Max Activating Token Index: 455

Text #13

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.6744. Min Act: -0.0000

Data Index: 182887 (C4 (Web Text))

Max Activating Token Index: 452

Text #14

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.6655. Min Act: -0.0000

Data Index: 883317 (C4 (Web Text))

Max Activating Token Index: 140

Text #15

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.6637. Min Act: -0.0000

Data Index: 1252578 (C4 (Web Text))

Max Activating Token Index: 213

Text #16

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.6084. Min Act: -0.0000

Data Index: 95445 (C4 (Web Text))

Max Activating Token Index: 83

Text #17

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.6413. Min Act: -0.0000

Data Index: 962963 (C4 (Web Text))

Max Activating Token Index: 156

Text #18

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.6293. Min Act: -0.0000

Data Index: 899906 (C4 (Web Text))

Max Activating Token Index: 744

Text #19

Max Range: 0.8181. Min Range: -0.8181

Max Act: 0.6251. Min Act: -0.0000

Data Index: 191144 (C4 (Web Text))

Max Activating Token Index: 334