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

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

Neuron 1803 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: 0.0197. Min Range: -0.0197

Max Act: 0.0197. Min Act: -0.0006

Data Index: 384310 (C4 (Web Text))

Max Activating Token Index: 713

Text #1

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0168. Min Act: -0.0006

Data Index: 1444387 (Python Code)

Max Activating Token Index: 101

Text #2

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0164. Min Act: -0.0006

Data Index: 1285338 (C4 (Web Text))

Max Activating Token Index: 358

Text #3

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0167. Min Act: -0.0006

Data Index: 933074 (C4 (Web Text))

Max Activating Token Index: 37

Text #4

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0150. Min Act: -0.0006

Data Index: 1394686 (Python Code)

Max Activating Token Index: 898

Text #5

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0151. Min Act: -0.0006

Data Index: 922949 (C4 (Web Text))

Max Activating Token Index: 345

Text #6

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0150. Min Act: -0.0006

Data Index: 223731 (C4 (Web Text))

Max Activating Token Index: 87

Text #7

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0145. Min Act: -0.0006

Data Index: 1423205 (Python Code)

Max Activating Token Index: 113

Text #8

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0151. Min Act: -0.0006

Data Index: 848116 (C4 (Web Text))

Max Activating Token Index: 269

Text #9

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0144. Min Act: -0.0006

Data Index: 781295 (C4 (Web Text))

Max Activating Token Index: 744

Text #10

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0149. Min Act: -0.0006

Data Index: 411107 (C4 (Web Text))

Max Activating Token Index: 588

Text #11

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0150. Min Act: -0.0006

Data Index: 1555536 (Python Code)

Max Activating Token Index: 408

Text #12

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0145. Min Act: -0.0006

Data Index: 1017467 (C4 (Web Text))

Max Activating Token Index: 514

Text #13

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0145. Min Act: -0.0006

Data Index: 1548439 (Python Code)

Max Activating Token Index: 124

Text #14

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0141. Min Act: -0.0006

Data Index: 1604493 (Python Code)

Max Activating Token Index: 533

Text #15

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0143. Min Act: -0.0006

Data Index: 1628548 (Python Code)

Max Activating Token Index: 214

Text #16

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0140. Min Act: -0.0006

Data Index: 374276 (C4 (Web Text))

Max Activating Token Index: 45

Text #17

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0146. Min Act: -0.0006

Data Index: 1470847 (Python Code)

Max Activating Token Index: 834

Text #18

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0142. Min Act: -0.0006

Data Index: 992059 (C4 (Web Text))

Max Activating Token Index: 224

Text #19

Max Range: 0.0197. Min Range: -0.0197

Max Act: 0.0138. Min Act: -0.0006

Data Index: 1454039 (Python Code)

Max Activating Token Index: 681