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

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

Neuron 3793 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.6961. Min Act: -0.0001

Data Index: 1406412 (Python Code)

Max Activating Token Index: 127

Text #1

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.6262. Min Act: -0.0001

Data Index: 295209 (C4 (Web Text))

Max Activating Token Index: 105

Text #2

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.7150. Min Act: -0.0001

Data Index: 176384 (C4 (Web Text))

Max Activating Token Index: 453

Text #3

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.3659. Min Act: -0.0001

Data Index: 164695 (C4 (Web Text))

Max Activating Token Index: 280

Text #4

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.4756. Min Act: -0.0002

Data Index: 506596 (C4 (Web Text))

Max Activating Token Index: 97

Text #5

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.4631. Min Act: -0.0002

Data Index: 86561 (C4 (Web Text))

Max Activating Token Index: 957

Text #6

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.4171. Min Act: -0.0002

Data Index: 211937 (C4 (Web Text))

Max Activating Token Index: 568

Text #7

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.4409. Min Act: -0.0001

Data Index: 799278 (C4 (Web Text))

Max Activating Token Index: 123

Text #8

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.2543. Min Act: -0.0002

Data Index: 355255 (C4 (Web Text))

Max Activating Token Index: 568

Text #9

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.4536. Min Act: -0.0001

Data Index: 462491 (C4 (Web Text))

Max Activating Token Index: 238

Text #10

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.3612. Min Act: -0.0002

Data Index: 1234051 (C4 (Web Text))

Max Activating Token Index: 137

Text #11

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.4091. Min Act: -0.0001

Data Index: 855163 (C4 (Web Text))

Max Activating Token Index: 122

Text #12

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.5464. Min Act: -0.0002

Data Index: 672866 (C4 (Web Text))

Max Activating Token Index: 768

Text #13

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.6006. Min Act: -0.0002

Data Index: 684106 (C4 (Web Text))

Max Activating Token Index: 646

Text #14

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.5951. Min Act: -0.0002

Data Index: 396503 (C4 (Web Text))

Max Activating Token Index: 678

Text #15

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.5102. Min Act: -0.0002

Data Index: 611303 (C4 (Web Text))

Max Activating Token Index: 302

Text #16

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.4790. Min Act: -0.0002

Data Index: 292631 (C4 (Web Text))

Max Activating Token Index: 822

Text #17

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.4771. Min Act: -0.0002

Data Index: 904849 (C4 (Web Text))

Max Activating Token Index: 78

Text #18

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.3505. Min Act: -0.0002

Data Index: 1193107 (C4 (Web Text))

Max Activating Token Index: 745

Text #19

Max Range: 1.7150. Min Range: -1.7150

Max Act: 1.3758. Min Act: -0.0002

Data Index: 231759 (C4 (Web Text))

Max Activating Token Index: 990