Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 780 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1164. Min Act: -0.0001

Data Index: 2339 (C4 (Web Text))

Max Activating Token Index: 307

Text #1

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1165. Min Act: -0.0001

Data Index: 1057140 (C4 (Web Text))

Max Activating Token Index: 564

Text #2

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1125. Min Act: -0.0001

Data Index: 979507 (C4 (Web Text))

Max Activating Token Index: 227

Text #3

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.0983. Min Act: -0.0001

Data Index: 818282 (C4 (Web Text))

Max Activating Token Index: 319

Text #4

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1135. Min Act: -0.0001

Data Index: 1248589 (C4 (Web Text))

Max Activating Token Index: 238

Text #5

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1089. Min Act: -0.0001

Data Index: 928144 (C4 (Web Text))

Max Activating Token Index: 435

Text #6

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1049. Min Act: -0.0001

Data Index: 874604 (C4 (Web Text))

Max Activating Token Index: 167

Text #7

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1086. Min Act: -0.0001

Data Index: 741016 (C4 (Web Text))

Max Activating Token Index: 699

Text #8

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1068. Min Act: -0.0001

Data Index: 223029 (C4 (Web Text))

Max Activating Token Index: 457

Text #9

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1054. Min Act: -0.0001

Data Index: 909335 (C4 (Web Text))

Max Activating Token Index: 199

Text #10

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1047. Min Act: -0.0001

Data Index: 495296 (C4 (Web Text))

Max Activating Token Index: 88

Text #11

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1084. Min Act: -0.0001

Data Index: 734223 (C4 (Web Text))

Max Activating Token Index: 82

Text #12

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.0987. Min Act: -0.0001

Data Index: 702042 (C4 (Web Text))

Max Activating Token Index: 604

Text #13

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.0995. Min Act: -0.0001

Data Index: 460482 (C4 (Web Text))

Max Activating Token Index: 566

Text #14

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1024. Min Act: -0.0001

Data Index: 1248616 (C4 (Web Text))

Max Activating Token Index: 139

Text #15

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.0977. Min Act: -0.0001

Data Index: 325341 (C4 (Web Text))

Max Activating Token Index: 633

Text #16

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1024. Min Act: -0.0001

Data Index: 146526 (C4 (Web Text))

Max Activating Token Index: 32

Text #17

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.0967. Min Act: -0.0001

Data Index: 1111635 (C4 (Web Text))

Max Activating Token Index: 333

Text #18

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.1008. Min Act: -0.0001

Data Index: 745015 (C4 (Web Text))

Max Activating Token Index: 788

Text #19

Max Range: 0.1165. Min Range: -0.1165

Max Act: 0.0936. Min Act: -0.0001

Data Index: 1303997 (C4 (Web Text))

Max Activating Token Index: 234