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

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

Neuron 386 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: 0.0216. Min Range: -0.0216

Max Act: 0.0216. Min Act: -0.0006

Data Index: 329306 (C4 (Web Text))

Max Activating Token Index: 164

Text #1

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0159. Min Act: -0.0006

Data Index: 1460957 (Python Code)

Max Activating Token Index: 150

Text #2

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0151. Min Act: -0.0006

Data Index: 396142 (C4 (Web Text))

Max Activating Token Index: 888

Text #3

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0152. Min Act: -0.0006

Data Index: 453613 (C4 (Web Text))

Max Activating Token Index: 916

Text #4

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0152. Min Act: -0.0006

Data Index: 556940 (C4 (Web Text))

Max Activating Token Index: 851

Text #5

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0148. Min Act: -0.0006

Data Index: 715428 (C4 (Web Text))

Max Activating Token Index: 409

Text #6

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0143. Min Act: -0.0006

Data Index: 797999 (C4 (Web Text))

Max Activating Token Index: 612

Text #7

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0138. Min Act: -0.0006

Data Index: 779082 (C4 (Web Text))

Max Activating Token Index: 594

Text #8

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0141. Min Act: -0.0006

Data Index: 1319114 (C4 (Web Text))

Max Activating Token Index: 741

Text #9

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0136. Min Act: -0.0006

Data Index: 1473071 (Python Code)

Max Activating Token Index: 325

Text #10

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0137. Min Act: -0.0006

Data Index: 162666 (C4 (Web Text))

Max Activating Token Index: 868

Text #11

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0133. Min Act: -0.0006

Data Index: 1326539 (C4 (Web Text))

Max Activating Token Index: 356

Text #12

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0133. Min Act: -0.0006

Data Index: 680371 (C4 (Web Text))

Max Activating Token Index: 409

Text #13

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0132. Min Act: -0.0006

Data Index: 955213 (C4 (Web Text))

Max Activating Token Index: 839

Text #14

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0135. Min Act: -0.0006

Data Index: 1459905 (Python Code)

Max Activating Token Index: 224

Text #15

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0133. Min Act: -0.0006

Data Index: 899555 (C4 (Web Text))

Max Activating Token Index: 453

Text #16

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0135. Min Act: -0.0006

Data Index: 1028293 (C4 (Web Text))

Max Activating Token Index: 304

Text #17

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0134. Min Act: -0.0006

Data Index: 1115569 (C4 (Web Text))

Max Activating Token Index: 281

Text #18

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0131. Min Act: -0.0006

Data Index: 1105348 (C4 (Web Text))

Max Activating Token Index: 26

Text #19

Max Range: 0.0216. Min Range: -0.0216

Max Act: 0.0137. Min Act: -0.0006

Data Index: 1258861 (C4 (Web Text))

Max Activating Token Index: 654