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

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

Neuron 2705 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: 0.8069. Min Range: -0.8069

Max Act: 0.7900. Min Act: -0.0001

Data Index: 653194 (C4 (Web Text))

Max Activating Token Index: 440

Text #1

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.8069. Min Act: -0.0001

Data Index: 44905 (C4 (Web Text))

Max Activating Token Index: 404

Text #2

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.7795. Min Act: -0.0001

Data Index: 1308601 (C4 (Web Text))

Max Activating Token Index: 838

Text #3

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.7342. Min Act: -0.0001

Data Index: 1186835 (C4 (Web Text))

Max Activating Token Index: 923

Text #4

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.7129. Min Act: -0.0001

Data Index: 1005504 (C4 (Web Text))

Max Activating Token Index: 658

Text #5

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.6153. Min Act: -0.0001

Data Index: 679175 (C4 (Web Text))

Max Activating Token Index: 659

Text #6

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.6531. Min Act: -0.0001

Data Index: 478335 (C4 (Web Text))

Max Activating Token Index: 394

Text #7

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.6390. Min Act: -0.0001

Data Index: 108886 (C4 (Web Text))

Max Activating Token Index: 836

Text #8

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.6161. Min Act: -0.0001

Data Index: 213472 (C4 (Web Text))

Max Activating Token Index: 103

Text #9

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.5970. Min Act: -0.0001

Data Index: 1008946 (C4 (Web Text))

Max Activating Token Index: 539

Text #10

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.6295. Min Act: -0.0001

Data Index: 900610 (C4 (Web Text))

Max Activating Token Index: 424

Text #11

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.6359. Min Act: -0.0001

Data Index: 1071389 (C4 (Web Text))

Max Activating Token Index: 255

Text #12

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.5981. Min Act: -0.0001

Data Index: 1426628 (Python Code)

Max Activating Token Index: 168

Text #13

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.5944. Min Act: -0.0001

Data Index: 941082 (C4 (Web Text))

Max Activating Token Index: 461

Text #14

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.5700. Min Act: -0.0001

Data Index: 979195 (C4 (Web Text))

Max Activating Token Index: 907

Text #15

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.5903. Min Act: -0.0000

Data Index: 1465101 (Python Code)

Max Activating Token Index: 335

Text #16

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.6299. Min Act: -0.0001

Data Index: 749756 (C4 (Web Text))

Max Activating Token Index: 342

Text #17

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.5789. Min Act: -0.0001

Data Index: 258115 (C4 (Web Text))

Max Activating Token Index: 567

Text #18

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.5694. Min Act: -0.0001

Data Index: 791898 (C4 (Web Text))

Max Activating Token Index: 862

Text #19

Max Range: 0.8069. Min Range: -0.8069

Max Act: 0.5894. Min Act: -0.0001

Data Index: 887302 (C4 (Web Text))

Max Activating Token Index: 23