Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4652 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.1186. Min Range: -1.1186

Max Act: 1.1080. Min Act: -0.0000

Data Index: 1471064 (Python Code)

Max Activating Token Index: 713

Text #1

Max Range: 1.1186. Min Range: -1.1186

Max Act: 1.1186. Min Act: -0.0000

Data Index: 1489439 (Python Code)

Max Activating Token Index: 457

Text #2

Max Range: 1.1186. Min Range: -1.1186

Max Act: 1.0252. Min Act: -0.0000

Data Index: 1111921 (C4 (Web Text))

Max Activating Token Index: 837

Text #3

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.9942. Min Act: -0.0000

Data Index: 1645703 (Python Code)

Max Activating Token Index: 915

Text #4

Max Range: 1.1186. Min Range: -1.1186

Max Act: 1.0165. Min Act: -0.0000

Data Index: 421552 (C4 (Web Text))

Max Activating Token Index: 954

Text #5

Max Range: 1.1186. Min Range: -1.1186

Max Act: 1.0473. Min Act: -0.0000

Data Index: 539727 (C4 (Web Text))

Max Activating Token Index: 362

Text #6

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.9580. Min Act: -0.0000

Data Index: 242872 (C4 (Web Text))

Max Activating Token Index: 948

Text #7

Max Range: 1.1186. Min Range: -1.1186

Max Act: 1.0282. Min Act: -0.0000

Data Index: 1520438 (Python Code)

Max Activating Token Index: 209

Text #8

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.9726. Min Act: -0.0000

Data Index: 1216748 (C4 (Web Text))

Max Activating Token Index: 997

Text #9

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.9695. Min Act: -0.0000

Data Index: 234866 (C4 (Web Text))

Max Activating Token Index: 22

Text #10

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.9608. Min Act: -0.0000

Data Index: 482022 (C4 (Web Text))

Max Activating Token Index: 64

Text #11

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.9245. Min Act: -0.0000

Data Index: 1653308 (Python Code)

Max Activating Token Index: 566

Text #12

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.9169. Min Act: -0.0000

Data Index: 1009226 (C4 (Web Text))

Max Activating Token Index: 703

Text #13

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.9437. Min Act: -0.0000

Data Index: 378384 (C4 (Web Text))

Max Activating Token Index: 438

Text #14

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.9099. Min Act: -0.0000

Data Index: 1231837 (C4 (Web Text))

Max Activating Token Index: 69

Text #15

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.9314. Min Act: -0.0000

Data Index: 433708 (C4 (Web Text))

Max Activating Token Index: 172

Text #16

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.9290. Min Act: -0.0000

Data Index: 326327 (C4 (Web Text))

Max Activating Token Index: 420

Text #17

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.8357. Min Act: -0.0000

Data Index: 1540989 (Python Code)

Max Activating Token Index: 20

Text #18

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.8860. Min Act: -0.0000

Data Index: 53779 (C4 (Web Text))

Max Activating Token Index: 738

Text #19

Max Range: 1.1186. Min Range: -1.1186

Max Act: 0.9116. Min Act: -0.0000

Data Index: 52286 (C4 (Web Text))

Max Activating Token Index: 485