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

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

Neuron 38 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: 0.0246. Min Range: -0.0246

Max Act: 0.0246. Min Act: -0.0006

Data Index: 310923 (C4 (Web Text))

Max Activating Token Index: 671

Text #1

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0242. Min Act: -0.0006

Data Index: 963989 (C4 (Web Text))

Max Activating Token Index: 960

Text #2

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0217. Min Act: -0.0006

Data Index: 926209 (C4 (Web Text))

Max Activating Token Index: 791

Text #3

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0217. Min Act: -0.0006

Data Index: 471856 (C4 (Web Text))

Max Activating Token Index: 219

Text #4

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0212. Min Act: -0.0006

Data Index: 746512 (C4 (Web Text))

Max Activating Token Index: 841

Text #5

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0203. Min Act: -0.0006

Data Index: 1584210 (Python Code)

Max Activating Token Index: 799

Text #6

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0205. Min Act: -0.0006

Data Index: 1323783 (C4 (Web Text))

Max Activating Token Index: 697

Text #7

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0213. Min Act: -0.0006

Data Index: 782178 (C4 (Web Text))

Max Activating Token Index: 842

Text #8

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0214. Min Act: -0.0006

Data Index: 195029 (C4 (Web Text))

Max Activating Token Index: 716

Text #9

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0214. Min Act: -0.0006

Data Index: 106580 (C4 (Web Text))

Max Activating Token Index: 726

Text #10

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0210. Min Act: -0.0006

Data Index: 794763 (C4 (Web Text))

Max Activating Token Index: 717

Text #11

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0207. Min Act: -0.0006

Data Index: 1491371 (Python Code)

Max Activating Token Index: 763

Text #12

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0198. Min Act: -0.0006

Data Index: 801380 (C4 (Web Text))

Max Activating Token Index: 362

Text #13

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0203. Min Act: -0.0006

Data Index: 178614 (C4 (Web Text))

Max Activating Token Index: 949

Text #14

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0202. Min Act: -0.0006

Data Index: 1008680 (C4 (Web Text))

Max Activating Token Index: 140

Text #15

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0204. Min Act: -0.0006

Data Index: 1147227 (C4 (Web Text))

Max Activating Token Index: 548

Text #16

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0196. Min Act: -0.0006

Data Index: 1155975 (C4 (Web Text))

Max Activating Token Index: 836

Text #17

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0199. Min Act: -0.0006

Data Index: 1291664 (C4 (Web Text))

Max Activating Token Index: 548

Text #18

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0202. Min Act: -0.0006

Data Index: 1185503 (C4 (Web Text))

Max Activating Token Index: 588

Text #19

Max Range: 0.0246. Min Range: -0.0246

Max Act: 0.0197. Min Act: -0.0006

Data Index: 1439864 (Python Code)

Max Activating Token Index: 888