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

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

Neuron 1995 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Click the toggle to see the full text!

Text #0

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.8462. Min Act: -0.0002

Data Index: 788796 (C4 (Web Text))

Max Activating Token Index: 719

Text #1

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.8640. Min Act: -0.0002

Data Index: 525533 (C4 (Web Text))

Max Activating Token Index: 22

Text #2

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.8155. Min Act: -0.0002

Data Index: 456881 (C4 (Web Text))

Max Activating Token Index: 26

Text #3

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7862. Min Act: -0.0002

Data Index: 526236 (C4 (Web Text))

Max Activating Token Index: 3

Text #4

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7712. Min Act: -0.0002

Data Index: 705149 (C4 (Web Text))

Max Activating Token Index: 501

Text #5

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7444. Min Act: -0.0002

Data Index: 430523 (C4 (Web Text))

Max Activating Token Index: 302

Text #6

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7527. Min Act: -0.0002

Data Index: 66576 (C4 (Web Text))

Max Activating Token Index: 36

Text #7

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7436. Min Act: -0.0003

Data Index: 396626 (C4 (Web Text))

Max Activating Token Index: 4

Text #8

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7584. Min Act: -0.0002

Data Index: 436549 (C4 (Web Text))

Max Activating Token Index: 641

Text #9

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7408. Min Act: -0.0002

Data Index: 110350 (C4 (Web Text))

Max Activating Token Index: 755

Text #10

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7378. Min Act: -0.0002

Data Index: 841803 (C4 (Web Text))

Max Activating Token Index: 8

Text #11

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7252. Min Act: -0.0002

Data Index: 1311785 (C4 (Web Text))

Max Activating Token Index: 2

Text #12

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7221. Min Act: -0.0002

Data Index: 1398275 (Python Code)

Max Activating Token Index: 726

Text #13

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7092. Min Act: -0.0002

Data Index: 1190698 (C4 (Web Text))

Max Activating Token Index: 6

Text #14

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7242. Min Act: -0.0002

Data Index: 133345 (C4 (Web Text))

Max Activating Token Index: 691

Text #15

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7113. Min Act: -0.0002

Data Index: 787006 (C4 (Web Text))

Max Activating Token Index: 197

Text #16

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7101. Min Act: -0.0002

Data Index: 88186 (C4 (Web Text))

Max Activating Token Index: 1018

Text #17

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7107. Min Act: -0.0002

Data Index: 781775 (C4 (Web Text))

Max Activating Token Index: 29

Text #18

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7130. Min Act: -0.0002

Data Index: 408264 (C4 (Web Text))

Max Activating Token Index: 2

Text #19

Max Range: 0.8640. Min Range: -0.8640

Max Act: 0.7130. Min Act: -0.0002

Data Index: 239292 (C4 (Web Text))

Max Activating Token Index: 2