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

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

Neuron 5635 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.3551. Min Act: -0.0000

Data Index: 926836 (C4 (Web Text))

Max Activating Token Index: 701

Text #1

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.3382. Min Act: -0.0000

Data Index: 1022050 (C4 (Web Text))

Max Activating Token Index: 157

Text #2

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.3158. Min Act: -0.0000

Data Index: 184964 (C4 (Web Text))

Max Activating Token Index: 715

Text #3

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.3328. Min Act: -0.0000

Data Index: 191502 (C4 (Web Text))

Max Activating Token Index: 273

Text #4

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.3234. Min Act: -0.0000

Data Index: 503811 (C4 (Web Text))

Max Activating Token Index: 277

Text #5

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.3153. Min Act: -0.0000

Data Index: 514800 (C4 (Web Text))

Max Activating Token Index: 201

Text #6

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.3106. Min Act: -0.0000

Data Index: 311771 (C4 (Web Text))

Max Activating Token Index: 549

Text #7

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.3038. Min Act: -0.0000

Data Index: 967880 (C4 (Web Text))

Max Activating Token Index: 481

Text #8

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.2865. Min Act: -0.0000

Data Index: 473531 (C4 (Web Text))

Max Activating Token Index: 192

Text #9

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.2606. Min Act: -0.0000

Data Index: 388104 (C4 (Web Text))

Max Activating Token Index: 823

Text #10

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.2786. Min Act: -0.0000

Data Index: 1468833 (Python Code)

Max Activating Token Index: 17

Text #11

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.2779. Min Act: -0.0000

Data Index: 724246 (C4 (Web Text))

Max Activating Token Index: 183

Text #12

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.2862. Min Act: -0.0000

Data Index: 1319861 (C4 (Web Text))

Max Activating Token Index: 100

Text #13

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.2569. Min Act: -0.0000

Data Index: 890001 (C4 (Web Text))

Max Activating Token Index: 683

Text #14

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.2663. Min Act: -0.0000

Data Index: 1040393 (C4 (Web Text))

Max Activating Token Index: 214

Text #15

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.2734. Min Act: 0.0000

Data Index: 1496556 (Python Code)

Max Activating Token Index: 18

Text #16

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.2825. Min Act: 0.0000

Data Index: 1373376 (Python Code)

Max Activating Token Index: 18

Text #17

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.2604. Min Act: -0.0000

Data Index: 36757 (C4 (Web Text))

Max Activating Token Index: 245

Text #18

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.2637. Min Act: -0.0000

Data Index: 1552303 (Python Code)

Max Activating Token Index: 15

Text #19

Max Range: 0.3551. Min Range: -0.3551

Max Act: 0.2641. Min Act: -0.0000

Data Index: 1210685 (C4 (Web Text))

Max Activating Token Index: 325