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

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

Neuron 275 in Layer 1

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: 922746 (C4 (Web Text))

Max Activating Token Index: 539

Text #1

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0161. Min Act: -0.0006

Data Index: 1179093 (C4 (Web Text))

Max Activating Token Index: 261

Text #2

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0148. Min Act: -0.0006

Data Index: 1417599 (Python Code)

Max Activating Token Index: 308

Text #3

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0152. Min Act: -0.0006

Data Index: 1044438 (C4 (Web Text))

Max Activating Token Index: 25

Text #4

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0147. Min Act: -0.0006

Data Index: 1393019 (Python Code)

Max Activating Token Index: 904

Text #5

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0144. Min Act: -0.0006

Data Index: 913134 (C4 (Web Text))

Max Activating Token Index: 397

Text #6

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0156. Min Act: -0.0006

Data Index: 1280497 (C4 (Web Text))

Max Activating Token Index: 332

Text #7

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0148. Min Act: -0.0006

Data Index: 326417 (C4 (Web Text))

Max Activating Token Index: 571

Text #8

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0135. Min Act: -0.0006

Data Index: 635024 (C4 (Web Text))

Max Activating Token Index: 954

Text #9

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0141. Min Act: -0.0006

Data Index: 181603 (C4 (Web Text))

Max Activating Token Index: 296

Text #10

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0125. Min Act: -0.0006

Data Index: 1157718 (C4 (Web Text))

Max Activating Token Index: 239

Text #11

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0147. Min Act: -0.0005

Data Index: 1181256 (C4 (Web Text))

Max Activating Token Index: 284

Text #12

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0143. Min Act: -0.0006

Data Index: 941560 (C4 (Web Text))

Max Activating Token Index: 568

Text #13

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0145. Min Act: -0.0006

Data Index: 983663 (C4 (Web Text))

Max Activating Token Index: 488

Text #14

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0158. Min Act: -0.0006

Data Index: 820324 (C4 (Web Text))

Max Activating Token Index: 489

Text #15

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0138. Min Act: -0.0006

Data Index: 337759 (C4 (Web Text))

Max Activating Token Index: 837

Text #16

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0148. Min Act: -0.0006

Data Index: 875704 (C4 (Web Text))

Max Activating Token Index: 951

Text #17

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0147. Min Act: -0.0006

Data Index: 658219 (C4 (Web Text))

Max Activating Token Index: 437

Text #18

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0137. Min Act: -0.0006

Data Index: 1412165 (Python Code)

Max Activating Token Index: 740

Text #19

Max Range: 0.0168. Min Range: -0.0168

Max Act: 0.0140. Min Act: -0.0006

Data Index: 612893 (C4 (Web Text))

Max Activating Token Index: 241