Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 23 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.9916. Min Act: -0.0004

Data Index: 418503 (C4 (Web Text))

Max Activating Token Index: 659

Text #1

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.9379. Min Act: -0.0004

Data Index: 823216 (C4 (Web Text))

Max Activating Token Index: 90

Text #2

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.9090. Min Act: -0.0004

Data Index: 1150401 (C4 (Web Text))

Max Activating Token Index: 563

Text #3

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.8745. Min Act: -0.0004

Data Index: 146438 (C4 (Web Text))

Max Activating Token Index: 644

Text #4

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.7962. Min Act: -0.0004

Data Index: 261428 (C4 (Web Text))

Max Activating Token Index: 1009

Text #5

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.8260. Min Act: -0.0004

Data Index: 78750 (C4 (Web Text))

Max Activating Token Index: 484

Text #6

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.8249. Min Act: -0.0004

Data Index: 1024017 (C4 (Web Text))

Max Activating Token Index: 69

Text #7

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.7261. Min Act: -0.0004

Data Index: 1040102 (C4 (Web Text))

Max Activating Token Index: 507

Text #8

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.7931. Min Act: -0.0004

Data Index: 515734 (C4 (Web Text))

Max Activating Token Index: 552

Text #9

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.8600. Min Act: -0.0004

Data Index: 1312009 (C4 (Web Text))

Max Activating Token Index: 574

Text #10

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.8260. Min Act: -0.0004

Data Index: 663895 (C4 (Web Text))

Max Activating Token Index: 163

Text #11

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.8688. Min Act: -0.0004

Data Index: 33832 (C4 (Web Text))

Max Activating Token Index: 255

Text #12

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.8020. Min Act: -0.0004

Data Index: 631957 (C4 (Web Text))

Max Activating Token Index: 833

Text #13

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.8057. Min Act: -0.0004

Data Index: 1165344 (C4 (Web Text))

Max Activating Token Index: 682

Text #14

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.7744. Min Act: -0.0004

Data Index: 186968 (C4 (Web Text))

Max Activating Token Index: 477

Text #15

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.7659. Min Act: -0.0004

Data Index: 690663 (C4 (Web Text))

Max Activating Token Index: 781

Text #16

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.7646. Min Act: -0.0004

Data Index: 1081060 (C4 (Web Text))

Max Activating Token Index: 778

Text #17

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.7573. Min Act: -0.0004

Data Index: 1343342 (C4 (Web Text))

Max Activating Token Index: 991

Text #18

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.7797. Min Act: -0.0004

Data Index: 437849 (C4 (Web Text))

Max Activating Token Index: 407

Text #19

Max Range: 0.9916. Min Range: -0.9916

Max Act: 0.7283. Min Act: -0.0004

Data Index: 267904 (C4 (Web Text))

Max Activating Token Index: 205