Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3193 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0247. Min Act: -0.0003

Data Index: 736718 (C4 (Web Text))

Max Activating Token Index: 481

Text #1

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0228. Min Act: -0.0003

Data Index: 1504193 (Python Code)

Max Activating Token Index: 653

Text #2

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0221. Min Act: -0.0003

Data Index: 526184 (C4 (Web Text))

Max Activating Token Index: 681

Text #3

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0211. Min Act: -0.0003

Data Index: 1120719 (C4 (Web Text))

Max Activating Token Index: 675

Text #4

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0231. Min Act: -0.0003

Data Index: 1371486 (Python Code)

Max Activating Token Index: 887

Text #5

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0239. Min Act: -0.0003

Data Index: 187762 (C4 (Web Text))

Max Activating Token Index: 841

Text #6

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0165. Min Act: -0.0003

Data Index: 1124999 (C4 (Web Text))

Max Activating Token Index: 670

Text #7

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0197. Min Act: -0.0003

Data Index: 759834 (C4 (Web Text))

Max Activating Token Index: 604

Text #8

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0197. Min Act: -0.0003

Data Index: 1157900 (C4 (Web Text))

Max Activating Token Index: 418

Text #9

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0205. Min Act: -0.0003

Data Index: 730040 (C4 (Web Text))

Max Activating Token Index: 799

Text #10

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0205. Min Act: -0.0004

Data Index: 839291 (C4 (Web Text))

Max Activating Token Index: 368

Text #11

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0190. Min Act: -0.0003

Data Index: 313548 (C4 (Web Text))

Max Activating Token Index: 773

Text #12

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0202. Min Act: -0.0003

Data Index: 807393 (C4 (Web Text))

Max Activating Token Index: 783

Text #13

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0197. Min Act: -0.0003

Data Index: 1419238 (Python Code)

Max Activating Token Index: 100

Text #14

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0181. Min Act: -0.0003

Data Index: 941627 (C4 (Web Text))

Max Activating Token Index: 75

Text #15

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0201. Min Act: -0.0003

Data Index: 361244 (C4 (Web Text))

Max Activating Token Index: 739

Text #16

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0201. Min Act: -0.0003

Data Index: 996314 (C4 (Web Text))

Max Activating Token Index: 784

Text #17

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0190. Min Act: -0.0003

Data Index: 715012 (C4 (Web Text))

Max Activating Token Index: 769

Text #18

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0182. Min Act: -0.0003

Data Index: 1536109 (Python Code)

Max Activating Token Index: 325

Text #19

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0196. Min Act: -0.0003

Data Index: 412425 (C4 (Web Text))

Max Activating Token Index: 957