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

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

Neuron 4143 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: 8.7184. Min Range: -8.7184

Max Act: 8.7184. Min Act: -0.0001

Data Index: 930524 (C4 (Web Text))

Max Activating Token Index: 141

Text #1

Max Range: 8.7184. Min Range: -8.7184

Max Act: 8.3779. Min Act: -0.0001

Data Index: 152320 (C4 (Web Text))

Max Activating Token Index: 457

Text #2

Max Range: 8.7184. Min Range: -8.7184

Max Act: 8.3364. Min Act: -0.0001

Data Index: 698693 (C4 (Web Text))

Max Activating Token Index: 710

Text #3

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.8782. Min Act: -0.0001

Data Index: 132478 (C4 (Web Text))

Max Activating Token Index: 85

Text #4

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.6142. Min Act: -0.0001

Data Index: 1089892 (C4 (Web Text))

Max Activating Token Index: 731

Text #5

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.9457. Min Act: -0.0001

Data Index: 1271093 (C4 (Web Text))

Max Activating Token Index: 263

Text #6

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.3854. Min Act: -0.0001

Data Index: 253630 (C4 (Web Text))

Max Activating Token Index: 1004

Text #7

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.5525. Min Act: -0.0001

Data Index: 1127672 (C4 (Web Text))

Max Activating Token Index: 213

Text #8

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.2793. Min Act: -0.0001

Data Index: 207130 (C4 (Web Text))

Max Activating Token Index: 548

Text #9

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.7332. Min Act: -0.0001

Data Index: 259230 (C4 (Web Text))

Max Activating Token Index: 298

Text #10

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.3820. Min Act: -0.0001

Data Index: 946241 (C4 (Web Text))

Max Activating Token Index: 405

Text #11

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.7027. Min Act: -0.0001

Data Index: 961490 (C4 (Web Text))

Max Activating Token Index: 131

Text #12

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.1443. Min Act: -0.0001

Data Index: 262395 (C4 (Web Text))

Max Activating Token Index: 942

Text #13

Max Range: 8.7184. Min Range: -8.7184

Max Act: 8.1630. Min Act: -0.0001

Data Index: 1353918 (C4 (Web Text))

Max Activating Token Index: 584

Text #14

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.6512. Min Act: -0.0001

Data Index: 980841 (C4 (Web Text))

Max Activating Token Index: 736

Text #15

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.1168. Min Act: -0.0001

Data Index: 487962 (C4 (Web Text))

Max Activating Token Index: 550

Text #16

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.1000. Min Act: -0.0001

Data Index: 291992 (C4 (Web Text))

Max Activating Token Index: 320

Text #17

Max Range: 8.7184. Min Range: -8.7184

Max Act: 6.9205. Min Act: -0.0001

Data Index: 33182 (C4 (Web Text))

Max Activating Token Index: 547

Text #18

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.3745. Min Act: -0.0001

Data Index: 565932 (C4 (Web Text))

Max Activating Token Index: 705

Text #19

Max Range: 8.7184. Min Range: -8.7184

Max Act: 7.1333. Min Act: -0.0001

Data Index: 383618 (C4 (Web Text))

Max Activating Token Index: 856