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

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

Neuron 900 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.3196. Min Range: -1.3196

Max Act: 1.2910. Min Act: -0.0001

Data Index: 755125 (C4 (Web Text))

Max Activating Token Index: 668

Text #1

Max Range: 1.3196. Min Range: -1.3196

Max Act: 1.3196. Min Act: -0.0001

Data Index: 182711 (C4 (Web Text))

Max Activating Token Index: 30

Text #2

Max Range: 1.3196. Min Range: -1.3196

Max Act: 1.2555. Min Act: -0.0001

Data Index: 217037 (C4 (Web Text))

Max Activating Token Index: 4

Text #3

Max Range: 1.3196. Min Range: -1.3196

Max Act: 1.1918. Min Act: -0.0001

Data Index: 1021174 (C4 (Web Text))

Max Activating Token Index: 58

Text #4

Max Range: 1.3196. Min Range: -1.3196

Max Act: 1.1203. Min Act: -0.0001

Data Index: 1148487 (C4 (Web Text))

Max Activating Token Index: 5

Text #5

Max Range: 1.3196. Min Range: -1.3196

Max Act: 1.1408. Min Act: -0.0001

Data Index: 1239097 (C4 (Web Text))

Max Activating Token Index: 824

Text #6

Max Range: 1.3196. Min Range: -1.3196

Max Act: 1.0866. Min Act: -0.0001

Data Index: 87490 (C4 (Web Text))

Max Activating Token Index: 10

Text #7

Max Range: 1.3196. Min Range: -1.3196

Max Act: 1.0578. Min Act: -0.0001

Data Index: 1288907 (C4 (Web Text))

Max Activating Token Index: 7

Text #8

Max Range: 1.3196. Min Range: -1.3196

Max Act: 0.9440. Min Act: -0.0001

Data Index: 443307 (C4 (Web Text))

Max Activating Token Index: 685

Text #9

Max Range: 1.3196. Min Range: -1.3196

Max Act: 1.0028. Min Act: -0.0001

Data Index: 709996 (C4 (Web Text))

Max Activating Token Index: 13

Text #10

Max Range: 1.3196. Min Range: -1.3196

Max Act: 1.0271. Min Act: -0.0001

Data Index: 689978 (C4 (Web Text))

Max Activating Token Index: 900

Text #11

Max Range: 1.3196. Min Range: -1.3196

Max Act: 1.0721. Min Act: -0.0001

Data Index: 1284825 (C4 (Web Text))

Max Activating Token Index: 617

Text #12

Max Range: 1.3196. Min Range: -1.3196

Max Act: 0.9730. Min Act: -0.0001

Data Index: 1154776 (C4 (Web Text))

Max Activating Token Index: 58

Text #13

Max Range: 1.3196. Min Range: -1.3196

Max Act: 0.9784. Min Act: -0.0001

Data Index: 171831 (C4 (Web Text))

Max Activating Token Index: 690

Text #14

Max Range: 1.3196. Min Range: -1.3196

Max Act: 0.9789. Min Act: -0.0001

Data Index: 318510 (C4 (Web Text))

Max Activating Token Index: 458

Text #15

Max Range: 1.3196. Min Range: -1.3196

Max Act: 0.9683. Min Act: -0.0001

Data Index: 992317 (C4 (Web Text))

Max Activating Token Index: 113

Text #16

Max Range: 1.3196. Min Range: -1.3196

Max Act: 0.9459. Min Act: -0.0001

Data Index: 506578 (C4 (Web Text))

Max Activating Token Index: 5

Text #17

Max Range: 1.3196. Min Range: -1.3196

Max Act: 0.9104. Min Act: -0.0001

Data Index: 40856 (C4 (Web Text))

Max Activating Token Index: 196

Text #18

Max Range: 1.3196. Min Range: -1.3196

Max Act: 0.9995. Min Act: -0.0001

Data Index: 960018 (C4 (Web Text))

Max Activating Token Index: 80

Text #19

Max Range: 1.3196. Min Range: -1.3196

Max Act: 0.9409. Min Act: -0.0001

Data Index: 1308221 (C4 (Web Text))

Max Activating Token Index: 147