Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3892 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.6325. Min Act: -0.0001

Data Index: 590479 (C4 (Web Text))

Max Activating Token Index: 51

Text #1

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.6221. Min Act: -0.0001

Data Index: 779524 (C4 (Web Text))

Max Activating Token Index: 160

Text #2

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.5448. Min Act: -0.0001

Data Index: 1297602 (C4 (Web Text))

Max Activating Token Index: 397

Text #3

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.5573. Min Act: -0.0001

Data Index: 1060069 (C4 (Web Text))

Max Activating Token Index: 43

Text #4

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.5449. Min Act: -0.0001

Data Index: 144722 (C4 (Web Text))

Max Activating Token Index: 117

Text #5

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.4687. Min Act: -0.0001

Data Index: 481134 (C4 (Web Text))

Max Activating Token Index: 922

Text #6

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.5216. Min Act: -0.0001

Data Index: 1038830 (C4 (Web Text))

Max Activating Token Index: 383

Text #7

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.5338. Min Act: -0.0002

Data Index: 262984 (C4 (Web Text))

Max Activating Token Index: 127

Text #8

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.4549. Min Act: -0.0001

Data Index: 939581 (C4 (Web Text))

Max Activating Token Index: 312

Text #9

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.5091. Min Act: -0.0001

Data Index: 294304 (C4 (Web Text))

Max Activating Token Index: 1007

Text #10

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.4791. Min Act: -0.0001

Data Index: 418703 (C4 (Web Text))

Max Activating Token Index: 594

Text #11

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.4432. Min Act: -0.0001

Data Index: 535013 (C4 (Web Text))

Max Activating Token Index: 294

Text #12

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.5003. Min Act: -0.0001

Data Index: 294657 (C4 (Web Text))

Max Activating Token Index: 494

Text #13

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.3876. Min Act: -0.0001

Data Index: 1259662 (C4 (Web Text))

Max Activating Token Index: 138

Text #14

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.4252. Min Act: -0.0001

Data Index: 1171668 (C4 (Web Text))

Max Activating Token Index: 862

Text #15

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.4387. Min Act: -0.0001

Data Index: 300127 (C4 (Web Text))

Max Activating Token Index: 543

Text #16

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.4467. Min Act: -0.0001

Data Index: 787768 (C4 (Web Text))

Max Activating Token Index: 911

Text #17

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.4292. Min Act: -0.0001

Data Index: 479723 (C4 (Web Text))

Max Activating Token Index: 24

Text #18

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.4612. Min Act: -0.0002

Data Index: 48161 (C4 (Web Text))

Max Activating Token Index: 990

Text #19

Max Range: 0.6325. Min Range: -0.6325

Max Act: 0.4230. Min Act: -0.0001

Data Index: 1066825 (C4 (Web Text))

Max Activating Token Index: 46