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

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

Neuron 4480 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 6.2250. Min Range: -6.2250

Max Act: 6.2250. Min Act: -0.0000

Data Index: 383280 (C4 (Web Text))

Max Activating Token Index: 976

Text #1

Max Range: 6.2250. Min Range: -6.2250

Max Act: 5.3693. Min Act: -0.0000

Data Index: 629711 (C4 (Web Text))

Max Activating Token Index: 239

Text #2

Max Range: 6.2250. Min Range: -6.2250

Max Act: 5.4677. Min Act: -0.0000

Data Index: 301341 (C4 (Web Text))

Max Activating Token Index: 776

Text #3

Max Range: 6.2250. Min Range: -6.2250

Max Act: 5.6627. Min Act: -0.0000

Data Index: 1044304 (C4 (Web Text))

Max Activating Token Index: 565

Text #4

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.5068. Min Act: -0.0000

Data Index: 461644 (C4 (Web Text))

Max Activating Token Index: 42

Text #5

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.8325. Min Act: -0.0000

Data Index: 654730 (C4 (Web Text))

Max Activating Token Index: 106

Text #6

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.6135. Min Act: -0.0000

Data Index: 77768 (C4 (Web Text))

Max Activating Token Index: 812

Text #7

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.7018. Min Act: -0.0000

Data Index: 240729 (C4 (Web Text))

Max Activating Token Index: 34

Text #8

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.6568. Min Act: -0.0000

Data Index: 1182350 (C4 (Web Text))

Max Activating Token Index: 931

Text #9

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.3489. Min Act: -0.0000

Data Index: 870885 (C4 (Web Text))

Max Activating Token Index: 414

Text #10

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.2769. Min Act: -0.0000

Data Index: 51223 (C4 (Web Text))

Max Activating Token Index: 419

Text #11

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.4188. Min Act: -0.0000

Data Index: 1084073 (C4 (Web Text))

Max Activating Token Index: 175

Text #12

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.6215. Min Act: -0.0000

Data Index: 169223 (C4 (Web Text))

Max Activating Token Index: 1004

Text #13

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.4133. Min Act: -0.0000

Data Index: 72238 (C4 (Web Text))

Max Activating Token Index: 191

Text #14

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.0030. Min Act: -0.0000

Data Index: 631620 (C4 (Web Text))

Max Activating Token Index: 652

Text #15

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.1232. Min Act: -0.0000

Data Index: 372055 (C4 (Web Text))

Max Activating Token Index: 336

Text #16

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.0346. Min Act: -0.0000

Data Index: 403172 (C4 (Web Text))

Max Activating Token Index: 416

Text #17

Max Range: 6.2250. Min Range: -6.2250

Max Act: 3.6909. Min Act: -0.0000

Data Index: 18546 (C4 (Web Text))

Max Activating Token Index: 647

Text #18

Max Range: 6.2250. Min Range: -6.2250

Max Act: 3.8305. Min Act: -0.0000

Data Index: 529409 (C4 (Web Text))

Max Activating Token Index: 15

Text #19

Max Range: 6.2250. Min Range: -6.2250

Max Act: 4.0363. Min Act: -0.0000

Data Index: 865201 (C4 (Web Text))

Max Activating Token Index: 587