Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3330 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.8788. Min Act: -0.0002

Data Index: 510636 (C4 (Web Text))

Max Activating Token Index: 76

Text #1

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.9167. Min Act: -0.0002

Data Index: 96545 (C4 (Web Text))

Max Activating Token Index: 270

Text #2

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.7235. Min Act: -0.0002

Data Index: 843633 (C4 (Web Text))

Max Activating Token Index: 42

Text #3

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.8101. Min Act: -0.0002

Data Index: 1184809 (C4 (Web Text))

Max Activating Token Index: 39

Text #4

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.8982. Min Act: -0.0002

Data Index: 653547 (C4 (Web Text))

Max Activating Token Index: 172

Text #5

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.6771. Min Act: -0.0002

Data Index: 917809 (C4 (Web Text))

Max Activating Token Index: 689

Text #6

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.6741. Min Act: -0.0003

Data Index: 885248 (C4 (Web Text))

Max Activating Token Index: 440

Text #7

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.6737. Min Act: -0.0003

Data Index: 889028 (C4 (Web Text))

Max Activating Token Index: 575

Text #8

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.5982. Min Act: -0.0002

Data Index: 691255 (C4 (Web Text))

Max Activating Token Index: 815

Text #9

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.5506. Min Act: -0.0002

Data Index: 1101274 (C4 (Web Text))

Max Activating Token Index: 15

Text #10

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.7099. Min Act: -0.0003

Data Index: 964749 (C4 (Web Text))

Max Activating Token Index: 244

Text #11

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.4877. Min Act: -0.0002

Data Index: 482756 (C4 (Web Text))

Max Activating Token Index: 20

Text #12

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.5913. Min Act: -0.0002

Data Index: 141518 (C4 (Web Text))

Max Activating Token Index: 44

Text #13

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.5961. Min Act: -0.0003

Data Index: 157498 (C4 (Web Text))

Max Activating Token Index: 96

Text #14

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.5241. Min Act: -0.0002

Data Index: 415627 (C4 (Web Text))

Max Activating Token Index: 50

Text #15

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.4077. Min Act: -0.0002

Data Index: 1191740 (C4 (Web Text))

Max Activating Token Index: 735

Text #16

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.4262. Min Act: -0.0002

Data Index: 1052151 (C4 (Web Text))

Max Activating Token Index: 54

Text #17

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.4616. Min Act: -0.0002

Data Index: 1052659 (C4 (Web Text))

Max Activating Token Index: 362

Text #18

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.4160. Min Act: -0.0002

Data Index: 208675 (C4 (Web Text))

Max Activating Token Index: 92

Text #19

Max Range: 1.9167. Min Range: -1.9167

Max Act: 1.4061. Min Act: -0.0003

Data Index: 398275 (C4 (Web Text))

Max Activating Token Index: 582