Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4108 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.7898. Min Act: -0.0001

Data Index: 1484610 (Python Code)

Max Activating Token Index: 762

Text #1

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.7871. Min Act: -0.0001

Data Index: 1307700 (C4 (Web Text))

Max Activating Token Index: 838

Text #2

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.8665. Min Act: -0.0001

Data Index: 1377139 (Python Code)

Max Activating Token Index: 1014

Text #3

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.6100. Min Act: -0.0001

Data Index: 129538 (C4 (Web Text))

Max Activating Token Index: 189

Text #4

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.4576. Min Act: -0.0001

Data Index: 639992 (C4 (Web Text))

Max Activating Token Index: 233

Text #5

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.4666. Min Act: -0.0001

Data Index: 470028 (C4 (Web Text))

Max Activating Token Index: 291

Text #6

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.5758. Min Act: -0.0001

Data Index: 235452 (C4 (Web Text))

Max Activating Token Index: 992

Text #7

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.4017. Min Act: -0.0001

Data Index: 817538 (C4 (Web Text))

Max Activating Token Index: 247

Text #8

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.4572. Min Act: -0.0001

Data Index: 1282278 (C4 (Web Text))

Max Activating Token Index: 960

Text #9

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.4836. Min Act: -0.0001

Data Index: 42266 (C4 (Web Text))

Max Activating Token Index: 568

Text #10

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.4458. Min Act: -0.0001

Data Index: 786527 (C4 (Web Text))

Max Activating Token Index: 250

Text #11

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.2898. Min Act: -0.0001

Data Index: 297272 (C4 (Web Text))

Max Activating Token Index: 215

Text #12

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.3474. Min Act: -0.0001

Data Index: 506146 (C4 (Web Text))

Max Activating Token Index: 785

Text #13

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.4651. Min Act: -0.0001

Data Index: 449955 (C4 (Web Text))

Max Activating Token Index: 654

Text #14

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.4048. Min Act: -0.0001

Data Index: 91121 (C4 (Web Text))

Max Activating Token Index: 929

Text #15

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.3606. Min Act: -0.0001

Data Index: 29382 (C4 (Web Text))

Max Activating Token Index: 846

Text #16

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.2376. Min Act: -0.0001

Data Index: 1426823 (Python Code)

Max Activating Token Index: 864

Text #17

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.3031. Min Act: -0.0001

Data Index: 1155971 (C4 (Web Text))

Max Activating Token Index: 525

Text #18

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.2337. Min Act: -0.0001

Data Index: 272155 (C4 (Web Text))

Max Activating Token Index: 690

Text #19

Max Range: 1.8665. Min Range: -1.8665

Max Act: 1.3385. Min Act: -0.0001

Data Index: 479032 (C4 (Web Text))

Max Activating Token Index: 344