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

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

Neuron 4550 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3409. Min Range: -1.3409

Max Act: 1.3409. Min Act: -0.0001

Data Index: 76620 (C4 (Web Text))

Max Activating Token Index: 305

Text #1

Max Range: 1.3409. Min Range: -1.3409

Max Act: 1.1157. Min Act: -0.0001

Data Index: 1221027 (C4 (Web Text))

Max Activating Token Index: 730

Text #2

Max Range: 1.3409. Min Range: -1.3409

Max Act: 1.0744. Min Act: -0.0001

Data Index: 300893 (C4 (Web Text))

Max Activating Token Index: 57

Text #3

Max Range: 1.3409. Min Range: -1.3409

Max Act: 1.0698. Min Act: -0.0001

Data Index: 952816 (C4 (Web Text))

Max Activating Token Index: 10

Text #4

Max Range: 1.3409. Min Range: -1.3409

Max Act: 1.0829. Min Act: -0.0001

Data Index: 225944 (C4 (Web Text))

Max Activating Token Index: 396

Text #5

Max Range: 1.3409. Min Range: -1.3409

Max Act: 1.0837. Min Act: -0.0001

Data Index: 275516 (C4 (Web Text))

Max Activating Token Index: 529

Text #6

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.9837. Min Act: -0.0001

Data Index: 769589 (C4 (Web Text))

Max Activating Token Index: 406

Text #7

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.9524. Min Act: -0.0001

Data Index: 100019 (C4 (Web Text))

Max Activating Token Index: 151

Text #8

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.9795. Min Act: -0.0001

Data Index: 617049 (C4 (Web Text))

Max Activating Token Index: 140

Text #9

Max Range: 1.3409. Min Range: -1.3409

Max Act: 1.0192. Min Act: -0.0001

Data Index: 504846 (C4 (Web Text))

Max Activating Token Index: 920

Text #10

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.9857. Min Act: -0.0001

Data Index: 691279 (C4 (Web Text))

Max Activating Token Index: 234

Text #11

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.9633. Min Act: -0.0001

Data Index: 803700 (C4 (Web Text))

Max Activating Token Index: 136

Text #12

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.9424. Min Act: -0.0001

Data Index: 1290067 (C4 (Web Text))

Max Activating Token Index: 889

Text #13

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.9389. Min Act: -0.0001

Data Index: 532931 (C4 (Web Text))

Max Activating Token Index: 195

Text #14

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.8790. Min Act: -0.0001

Data Index: 913369 (C4 (Web Text))

Max Activating Token Index: 894

Text #15

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.8743. Min Act: -0.0001

Data Index: 1281816 (C4 (Web Text))

Max Activating Token Index: 913

Text #16

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.8498. Min Act: -0.0001

Data Index: 493411 (C4 (Web Text))

Max Activating Token Index: 659

Text #17

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.8776. Min Act: -0.0001

Data Index: 1123033 (C4 (Web Text))

Max Activating Token Index: 238

Text #18

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.9205. Min Act: -0.0001

Data Index: 400739 (C4 (Web Text))

Max Activating Token Index: 100

Text #19

Max Range: 1.3409. Min Range: -1.3409

Max Act: 0.9402. Min Act: -0.0001

Data Index: 404792 (C4 (Web Text))

Max Activating Token Index: 244