Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 656 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0168. Min Act: -0.0006

Data Index: 204564 (C4 (Web Text))

Max Activating Token Index: 18

Text #1

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0155. Min Act: -0.0006

Data Index: 853840 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0146. Min Act: -0.0005

Data Index: 644829 (C4 (Web Text))

Max Activating Token Index: 799

Text #3

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0145. Min Act: -0.0005

Data Index: 1358940 (C4 (Web Text))

Max Activating Token Index: 1017

Text #4

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0140. Min Act: -0.0007

Data Index: 720311 (C4 (Web Text))

Max Activating Token Index: 615

Text #5

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0136. Min Act: -0.0006

Data Index: 106308 (C4 (Web Text))

Max Activating Token Index: 772

Text #6

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0136. Min Act: -0.0007

Data Index: 744910 (C4 (Web Text))

Max Activating Token Index: 455

Text #7

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0132. Min Act: -0.0006

Data Index: 885144 (C4 (Web Text))

Max Activating Token Index: 738

Text #8

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0134. Min Act: -0.0006

Data Index: 1006836 (C4 (Web Text))

Max Activating Token Index: 43

Text #9

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0133. Min Act: -0.0007

Data Index: 596265 (C4 (Web Text))

Max Activating Token Index: 885

Text #10

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0133. Min Act: -0.0006

Data Index: 1197795 (C4 (Web Text))

Max Activating Token Index: 191

Text #11

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0133. Min Act: -0.0006

Data Index: 1239730 (C4 (Web Text))

Max Activating Token Index: 624

Text #12

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0136. Min Act: -0.0007

Data Index: 716872 (C4 (Web Text))

Max Activating Token Index: 741

Text #13

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0133. Min Act: -0.0007

Data Index: 909565 (C4 (Web Text))

Max Activating Token Index: 297

Text #14

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0129. Min Act: -0.0007

Data Index: 599177 (C4 (Web Text))

Max Activating Token Index: 748

Text #15

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0131. Min Act: -0.0007

Data Index: 1341072 (C4 (Web Text))

Max Activating Token Index: 271

Text #16

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0131. Min Act: -0.0006

Data Index: 331426 (C4 (Web Text))

Max Activating Token Index: 94

Text #17

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0130. Min Act: -0.0007

Data Index: 1277033 (C4 (Web Text))

Max Activating Token Index: 197

Text #18

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0127. Min Act: -0.0006

Data Index: 1134724 (C4 (Web Text))

Max Activating Token Index: 952

Text #19

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0130. Min Act: -0.0005

Data Index: 1268498 (C4 (Web Text))

Max Activating Token Index: 416