Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4847 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.7717. Min Act: -0.0002

Data Index: 750713 (C4 (Web Text))

Max Activating Token Index: 27

Text #1

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.7916. Min Act: -0.0002

Data Index: 788848 (C4 (Web Text))

Max Activating Token Index: 573

Text #2

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.7280. Min Act: -0.0002

Data Index: 1109827 (C4 (Web Text))

Max Activating Token Index: 362

Text #3

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.6475. Min Act: -0.0002

Data Index: 659651 (C4 (Web Text))

Max Activating Token Index: 715

Text #4

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.6548. Min Act: -0.0002

Data Index: 376599 (C4 (Web Text))

Max Activating Token Index: 640

Text #5

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.5976. Min Act: -0.0002

Data Index: 1350635 (C4 (Web Text))

Max Activating Token Index: 816

Text #6

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.5166. Min Act: -0.0002

Data Index: 572328 (C4 (Web Text))

Max Activating Token Index: 306

Text #7

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.5332. Min Act: -0.0002

Data Index: 1517360 (Python Code)

Max Activating Token Index: 684

Text #8

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.5269. Min Act: -0.0002

Data Index: 1313644 (C4 (Web Text))

Max Activating Token Index: 555

Text #9

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.4920. Min Act: -0.0002

Data Index: 814200 (C4 (Web Text))

Max Activating Token Index: 230

Text #10

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.5052. Min Act: -0.0002

Data Index: 766365 (C4 (Web Text))

Max Activating Token Index: 96

Text #11

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.5142. Min Act: -0.0002

Data Index: 1493881 (Python Code)

Max Activating Token Index: 280

Text #12

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.5140. Min Act: -0.0002

Data Index: 367306 (C4 (Web Text))

Max Activating Token Index: 491

Text #13

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.5234. Min Act: -0.0002

Data Index: 534829 (C4 (Web Text))

Max Activating Token Index: 390

Text #14

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.4889. Min Act: -0.0002

Data Index: 1224024 (C4 (Web Text))

Max Activating Token Index: 552

Text #15

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.4722. Min Act: -0.0002

Data Index: 1137355 (C4 (Web Text))

Max Activating Token Index: 56

Text #16

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.4695. Min Act: -0.0002

Data Index: 792463 (C4 (Web Text))

Max Activating Token Index: 721

Text #17

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.4730. Min Act: -0.0002

Data Index: 1399157 (Python Code)

Max Activating Token Index: 327

Text #18

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.4767. Min Act: -0.0002

Data Index: 1102283 (C4 (Web Text))

Max Activating Token Index: 106

Text #19

Max Range: 0.7916. Min Range: -0.7916

Max Act: 0.4768. Min Act: -0.0002

Data Index: 411361 (C4 (Web Text))

Max Activating Token Index: 918