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

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

Neuron 2667 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.1197. Min Act: -0.0001

Data Index: 11247 (C4 (Web Text))

Max Activating Token Index: 219

Text #1

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.1158. Min Act: -0.0001

Data Index: 622942 (C4 (Web Text))

Max Activating Token Index: 62

Text #2

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.1139. Min Act: -0.0000

Data Index: 231168 (C4 (Web Text))

Max Activating Token Index: 101

Text #3

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.1221. Min Act: -0.0001

Data Index: 588083 (C4 (Web Text))

Max Activating Token Index: 289

Text #4

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.1140. Min Act: -0.0001

Data Index: 326982 (C4 (Web Text))

Max Activating Token Index: 226

Text #5

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.1106. Min Act: -0.0001

Data Index: 813547 (C4 (Web Text))

Max Activating Token Index: 31

Text #6

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.1065. Min Act: -0.0000

Data Index: 632032 (C4 (Web Text))

Max Activating Token Index: 110

Text #7

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.1097. Min Act: -0.0001

Data Index: 107275 (C4 (Web Text))

Max Activating Token Index: 413

Text #8

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.1093. Min Act: -0.0001

Data Index: 453093 (C4 (Web Text))

Max Activating Token Index: 127

Text #9

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.1030. Min Act: -0.0001

Data Index: 847758 (C4 (Web Text))

Max Activating Token Index: 66

Text #10

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.0978. Min Act: -0.0001

Data Index: 269533 (C4 (Web Text))

Max Activating Token Index: 714

Text #11

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.0991. Min Act: -0.0001

Data Index: 1342901 (C4 (Web Text))

Max Activating Token Index: 353

Text #12

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.0988. Min Act: -0.0001

Data Index: 457599 (C4 (Web Text))

Max Activating Token Index: 457

Text #13

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.0946. Min Act: -0.0001

Data Index: 234402 (C4 (Web Text))

Max Activating Token Index: 338

Text #14

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.0966. Min Act: -0.0001

Data Index: 884749 (C4 (Web Text))

Max Activating Token Index: 442

Text #15

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.0957. Min Act: -0.0001

Data Index: 1283149 (C4 (Web Text))

Max Activating Token Index: 285

Text #16

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.1030. Min Act: -0.0001

Data Index: 1161940 (C4 (Web Text))

Max Activating Token Index: 943

Text #17

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.0918. Min Act: -0.0001

Data Index: 566008 (C4 (Web Text))

Max Activating Token Index: 934

Text #18

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.0949. Min Act: -0.0001

Data Index: 1103953 (C4 (Web Text))

Max Activating Token Index: 849

Text #19

Max Range: 0.1221. Min Range: -0.1221

Max Act: 0.0968. Min Act: -0.0001

Data Index: 406544 (C4 (Web Text))

Max Activating Token Index: 572