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

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

Neuron 532 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: 0.0159. Min Range: -0.0159

Max Act: 0.0158. Min Act: -0.0006

Data Index: 1551677 (Python Code)

Max Activating Token Index: 507

Text #1

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0159. Min Act: -0.0006

Data Index: 244552 (C4 (Web Text))

Max Activating Token Index: 516

Text #2

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0146. Min Act: -0.0006

Data Index: 1607534 (Python Code)

Max Activating Token Index: 484

Text #3

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0144. Min Act: -0.0006

Data Index: 1255821 (C4 (Web Text))

Max Activating Token Index: 606

Text #4

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0147. Min Act: -0.0006

Data Index: 872894 (C4 (Web Text))

Max Activating Token Index: 167

Text #5

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0144. Min Act: -0.0006

Data Index: 289697 (C4 (Web Text))

Max Activating Token Index: 537

Text #6

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0138. Min Act: -0.0006

Data Index: 1148266 (C4 (Web Text))

Max Activating Token Index: 269

Text #7

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0140. Min Act: -0.0006

Data Index: 1565272 (Python Code)

Max Activating Token Index: 548

Text #8

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0139. Min Act: -0.0006

Data Index: 1557817 (Python Code)

Max Activating Token Index: 618

Text #9

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0129. Min Act: -0.0005

Data Index: 277764 (C4 (Web Text))

Max Activating Token Index: 70

Text #10

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0133. Min Act: -0.0006

Data Index: 250004 (C4 (Web Text))

Max Activating Token Index: 705

Text #11

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0130. Min Act: -0.0006

Data Index: 238663 (C4 (Web Text))

Max Activating Token Index: 774

Text #12

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0131. Min Act: -0.0006

Data Index: 1405166 (Python Code)

Max Activating Token Index: 387

Text #13

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0131. Min Act: -0.0006

Data Index: 37414 (C4 (Web Text))

Max Activating Token Index: 394

Text #14

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0130. Min Act: -0.0006

Data Index: 291431 (C4 (Web Text))

Max Activating Token Index: 842

Text #15

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0127. Min Act: -0.0006

Data Index: 338939 (C4 (Web Text))

Max Activating Token Index: 94

Text #16

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0126. Min Act: -0.0006

Data Index: 1177971 (C4 (Web Text))

Max Activating Token Index: 625

Text #17

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0128. Min Act: -0.0006

Data Index: 557760 (C4 (Web Text))

Max Activating Token Index: 586

Text #18

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0130. Min Act: -0.0006

Data Index: 929469 (C4 (Web Text))

Max Activating Token Index: 858

Text #19

Max Range: 0.0159. Min Range: -0.0159

Max Act: 0.0126. Min Act: -0.0006

Data Index: 1552202 (Python Code)

Max Activating Token Index: 840