Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 552 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0250. Min Act: -0.0006

Data Index: 1481684 (Python Code)

Max Activating Token Index: 382

Text #1

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0228. Min Act: -0.0006

Data Index: 1585124 (Python Code)

Max Activating Token Index: 58

Text #2

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0234. Min Act: -0.0006

Data Index: 1583006 (Python Code)

Max Activating Token Index: 977

Text #3

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0203. Min Act: -0.0006

Data Index: 1435401 (Python Code)

Max Activating Token Index: 128

Text #4

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0208. Min Act: -0.0006

Data Index: 1295607 (C4 (Web Text))

Max Activating Token Index: 86

Text #5

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0198. Min Act: -0.0006

Data Index: 1103117 (C4 (Web Text))

Max Activating Token Index: 921

Text #6

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0202. Min Act: -0.0006

Data Index: 1482427 (Python Code)

Max Activating Token Index: 699

Text #7

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0207. Min Act: -0.0006

Data Index: 714912 (C4 (Web Text))

Max Activating Token Index: 55

Text #8

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0200. Min Act: -0.0006

Data Index: 1379739 (Python Code)

Max Activating Token Index: 668

Text #9

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0200. Min Act: -0.0006

Data Index: 903490 (C4 (Web Text))

Max Activating Token Index: 23

Text #10

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0200. Min Act: -0.0006

Data Index: 734461 (C4 (Web Text))

Max Activating Token Index: 20

Text #11

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0199. Min Act: -0.0006

Data Index: 894577 (C4 (Web Text))

Max Activating Token Index: 61

Text #12

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0195. Min Act: -0.0006

Data Index: 655853 (C4 (Web Text))

Max Activating Token Index: 297

Text #13

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0191. Min Act: -0.0006

Data Index: 1446498 (Python Code)

Max Activating Token Index: 427

Text #14

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0188. Min Act: -0.0006

Data Index: 1371543 (Python Code)

Max Activating Token Index: 712

Text #15

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0198. Min Act: -0.0006

Data Index: 79453 (C4 (Web Text))

Max Activating Token Index: 143

Text #16

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0188. Min Act: -0.0006

Data Index: 1508702 (Python Code)

Max Activating Token Index: 591

Text #17

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0188. Min Act: -0.0006

Data Index: 1399252 (Python Code)

Max Activating Token Index: 764

Text #18

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0183. Min Act: -0.0006

Data Index: 1361940 (Python Code)

Max Activating Token Index: 357

Text #19

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0191. Min Act: -0.0006

Data Index: 966222 (C4 (Web Text))

Max Activating Token Index: 23