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

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

Neuron 622 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: 1.3658. Min Range: -1.3658

Max Act: 1.3658. Min Act: -0.0001

Data Index: 934984 (C4 (Web Text))

Max Activating Token Index: 217

Text #1

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.2718. Min Act: -0.0001

Data Index: 1574231 (Python Code)

Max Activating Token Index: 318

Text #2

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.0874. Min Act: -0.0001

Data Index: 1441780 (Python Code)

Max Activating Token Index: 201

Text #3

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.2178. Min Act: -0.0001

Data Index: 603889 (C4 (Web Text))

Max Activating Token Index: 404

Text #4

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.1799. Min Act: -0.0001

Data Index: 464533 (C4 (Web Text))

Max Activating Token Index: 372

Text #5

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.2030. Min Act: -0.0001

Data Index: 992556 (C4 (Web Text))

Max Activating Token Index: 575

Text #6

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.1597. Min Act: -0.0001

Data Index: 1428872 (Python Code)

Max Activating Token Index: 785

Text #7

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.2161. Min Act: -0.0001

Data Index: 508419 (C4 (Web Text))

Max Activating Token Index: 541

Text #8

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.1076. Min Act: -0.0001

Data Index: 470315 (C4 (Web Text))

Max Activating Token Index: 676

Text #9

Max Range: 1.3658. Min Range: -1.3658

Max Act: 0.8661. Min Act: -0.0001

Data Index: 1566485 (Python Code)

Max Activating Token Index: 178

Text #10

Max Range: 1.3658. Min Range: -1.3658

Max Act: 0.8202. Min Act: -0.0001

Data Index: 111894 (C4 (Web Text))

Max Activating Token Index: 311

Text #11

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.1129. Min Act: -0.0001

Data Index: 1124054 (C4 (Web Text))

Max Activating Token Index: 444

Text #12

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.1060. Min Act: -0.0001

Data Index: 62371 (C4 (Web Text))

Max Activating Token Index: 767

Text #13

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.0275. Min Act: -0.0001

Data Index: 113282 (C4 (Web Text))

Max Activating Token Index: 214

Text #14

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.1065. Min Act: -0.0001

Data Index: 1546629 (Python Code)

Max Activating Token Index: 693

Text #15

Max Range: 1.3658. Min Range: -1.3658

Max Act: 0.9349. Min Act: -0.0001

Data Index: 1389158 (Python Code)

Max Activating Token Index: 296

Text #16

Max Range: 1.3658. Min Range: -1.3658

Max Act: 0.9781. Min Act: -0.0001

Data Index: 98511 (C4 (Web Text))

Max Activating Token Index: 511

Text #17

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.0157. Min Act: -0.0001

Data Index: 1189634 (C4 (Web Text))

Max Activating Token Index: 110

Text #18

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.0382. Min Act: -0.0001

Data Index: 871378 (C4 (Web Text))

Max Activating Token Index: 287

Text #19

Max Range: 1.3658. Min Range: -1.3658

Max Act: 1.0976. Min Act: -0.0001

Data Index: 1602835 (Python Code)

Max Activating Token Index: 670