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

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

Neuron 3689 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: 0.9898. Min Range: -0.9898

Max Act: 0.9898. Min Act: -0.0001

Data Index: 1479556 (Python Code)

Max Activating Token Index: 78

Text #1

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.9750. Min Act: -0.0002

Data Index: 377113 (C4 (Web Text))

Max Activating Token Index: 841

Text #2

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.9195. Min Act: -0.0002

Data Index: 502192 (C4 (Web Text))

Max Activating Token Index: 820

Text #3

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.9117. Min Act: -0.0002

Data Index: 878412 (C4 (Web Text))

Max Activating Token Index: 501

Text #4

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.8658. Min Act: -0.0001

Data Index: 234657 (C4 (Web Text))

Max Activating Token Index: 282

Text #5

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.8986. Min Act: -0.0001

Data Index: 1106975 (C4 (Web Text))

Max Activating Token Index: 439

Text #6

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.8310. Min Act: -0.0002

Data Index: 798678 (C4 (Web Text))

Max Activating Token Index: 806

Text #7

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.8816. Min Act: -0.0001

Data Index: 1277698 (C4 (Web Text))

Max Activating Token Index: 615

Text #8

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.8528. Min Act: -0.0002

Data Index: 945926 (C4 (Web Text))

Max Activating Token Index: 469

Text #9

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.9018. Min Act: -0.0002

Data Index: 1044532 (C4 (Web Text))

Max Activating Token Index: 341

Text #10

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.8846. Min Act: -0.0001

Data Index: 1045998 (C4 (Web Text))

Max Activating Token Index: 34

Text #11

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.8566. Min Act: -0.0001

Data Index: 939067 (C4 (Web Text))

Max Activating Token Index: 99

Text #12

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.8090. Min Act: -0.0001

Data Index: 585463 (C4 (Web Text))

Max Activating Token Index: 796

Text #13

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.8600. Min Act: -0.0001

Data Index: 1279023 (C4 (Web Text))

Max Activating Token Index: 721

Text #14

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.8408. Min Act: -0.0001

Data Index: 1292020 (C4 (Web Text))

Max Activating Token Index: 820

Text #15

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.7810. Min Act: -0.0002

Data Index: 426565 (C4 (Web Text))

Max Activating Token Index: 695

Text #16

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.7940. Min Act: -0.0002

Data Index: 217782 (C4 (Web Text))

Max Activating Token Index: 614

Text #17

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.8139. Min Act: -0.0002

Data Index: 194363 (C4 (Web Text))

Max Activating Token Index: 803

Text #18

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.7896. Min Act: -0.0002

Data Index: 1150232 (C4 (Web Text))

Max Activating Token Index: 302

Text #19

Max Range: 0.9898. Min Range: -0.9898

Max Act: 0.8060. Min Act: -0.0001

Data Index: 1166841 (C4 (Web Text))

Max Activating Token Index: 880