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

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

Neuron 3036 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.6911. Min Act: -0.0001

Data Index: 829 (C4 (Web Text))

Max Activating Token Index: 324

Text #1

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.5964. Min Act: -0.0001

Data Index: 142694 (C4 (Web Text))

Max Activating Token Index: 48

Text #2

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.5867. Min Act: -0.0001

Data Index: 361174 (C4 (Web Text))

Max Activating Token Index: 384

Text #3

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.5332. Min Act: -0.0001

Data Index: 1246558 (C4 (Web Text))

Max Activating Token Index: 360

Text #4

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4928. Min Act: -0.0001

Data Index: 92526 (C4 (Web Text))

Max Activating Token Index: 46

Text #5

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.5661. Min Act: -0.0001

Data Index: 1176235 (C4 (Web Text))

Max Activating Token Index: 741

Text #6

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4766. Min Act: -0.0001

Data Index: 764248 (C4 (Web Text))

Max Activating Token Index: 376

Text #7

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4496. Min Act: -0.0001

Data Index: 613454 (C4 (Web Text))

Max Activating Token Index: 288

Text #8

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4652. Min Act: -0.0001

Data Index: 1010249 (C4 (Web Text))

Max Activating Token Index: 18

Text #9

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4829. Min Act: -0.0001

Data Index: 272478 (C4 (Web Text))

Max Activating Token Index: 54

Text #10

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4807. Min Act: -0.0001

Data Index: 776549 (C4 (Web Text))

Max Activating Token Index: 403

Text #11

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4946. Min Act: -0.0001

Data Index: 742357 (C4 (Web Text))

Max Activating Token Index: 29

Text #12

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4624. Min Act: -0.0001

Data Index: 517842 (C4 (Web Text))

Max Activating Token Index: 118

Text #13

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4766. Min Act: -0.0001

Data Index: 190895 (C4 (Web Text))

Max Activating Token Index: 145

Text #14

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4756. Min Act: -0.0001

Data Index: 365254 (C4 (Web Text))

Max Activating Token Index: 173

Text #15

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4884. Min Act: -0.0001

Data Index: 948835 (C4 (Web Text))

Max Activating Token Index: 21

Text #16

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4897. Min Act: -0.0001

Data Index: 1134455 (C4 (Web Text))

Max Activating Token Index: 965

Text #17

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4462. Min Act: -0.0001

Data Index: 1358700 (C4 (Web Text))

Max Activating Token Index: 1000

Text #18

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4692. Min Act: -0.0001

Data Index: 1194920 (C4 (Web Text))

Max Activating Token Index: 904

Text #19

Max Range: 0.6911. Min Range: -0.6911

Max Act: 0.4648. Min Act: -0.0001

Data Index: 1093047 (C4 (Web Text))

Max Activating Token Index: 823