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

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

Neuron 1570 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: 0.1260. Min Range: -0.1260

Max Act: 0.1260. Min Act: -0.0002

Data Index: 783189 (C4 (Web Text))

Max Activating Token Index: 398

Text #1

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1245. Min Act: -0.0002

Data Index: 604824 (C4 (Web Text))

Max Activating Token Index: 614

Text #2

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1139. Min Act: -0.0002

Data Index: 622794 (C4 (Web Text))

Max Activating Token Index: 134

Text #3

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1105. Min Act: -0.0002

Data Index: 803568 (C4 (Web Text))

Max Activating Token Index: 657

Text #4

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1107. Min Act: -0.0002

Data Index: 625917 (C4 (Web Text))

Max Activating Token Index: 642

Text #5

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1210. Min Act: -0.0002

Data Index: 278365 (C4 (Web Text))

Max Activating Token Index: 299

Text #6

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1103. Min Act: -0.0002

Data Index: 1163816 (C4 (Web Text))

Max Activating Token Index: 322

Text #7

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1145. Min Act: -0.0002

Data Index: 1224041 (C4 (Web Text))

Max Activating Token Index: 577

Text #8

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1146. Min Act: -0.0002

Data Index: 817559 (C4 (Web Text))

Max Activating Token Index: 367

Text #9

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1136. Min Act: -0.0002

Data Index: 313548 (C4 (Web Text))

Max Activating Token Index: 345

Text #10

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1088. Min Act: -0.0002

Data Index: 640414 (C4 (Web Text))

Max Activating Token Index: 569

Text #11

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1060. Min Act: -0.0002

Data Index: 225956 (C4 (Web Text))

Max Activating Token Index: 271

Text #12

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1104. Min Act: -0.0002

Data Index: 822856 (C4 (Web Text))

Max Activating Token Index: 533

Text #13

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1111. Min Act: -0.0002

Data Index: 78420 (C4 (Web Text))

Max Activating Token Index: 721

Text #14

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1083. Min Act: -0.0002

Data Index: 636927 (C4 (Web Text))

Max Activating Token Index: 184

Text #15

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1130. Min Act: -0.0002

Data Index: 94069 (C4 (Web Text))

Max Activating Token Index: 242

Text #16

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1045. Min Act: -0.0002

Data Index: 748016 (C4 (Web Text))

Max Activating Token Index: 800

Text #17

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1080. Min Act: -0.0002

Data Index: 934181 (C4 (Web Text))

Max Activating Token Index: 363

Text #18

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1056. Min Act: -0.0002

Data Index: 502098 (C4 (Web Text))

Max Activating Token Index: 40

Text #19

Max Range: 0.1260. Min Range: -0.1260

Max Act: 0.1099. Min Act: -0.0002

Data Index: 832122 (C4 (Web Text))

Max Activating Token Index: 451