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

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

Neuron 2517 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: 0.1782. Min Range: -0.1782

Max Act: 0.1782. Min Act: -0.0001

Data Index: 959007 (C4 (Web Text))

Max Activating Token Index: 282

Text #1

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1454. Min Act: -0.0001

Data Index: 584539 (C4 (Web Text))

Max Activating Token Index: 750

Text #2

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1478. Min Act: -0.0001

Data Index: 224384 (C4 (Web Text))

Max Activating Token Index: 219

Text #3

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1500. Min Act: -0.0001

Data Index: 978328 (C4 (Web Text))

Max Activating Token Index: 265

Text #4

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1514. Min Act: -0.0001

Data Index: 308334 (C4 (Web Text))

Max Activating Token Index: 654

Text #5

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1356. Min Act: -0.0001

Data Index: 719542 (C4 (Web Text))

Max Activating Token Index: 76

Text #6

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1496. Min Act: -0.0001

Data Index: 1137427 (C4 (Web Text))

Max Activating Token Index: 347

Text #7

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1361. Min Act: -0.0001

Data Index: 298562 (C4 (Web Text))

Max Activating Token Index: 92

Text #8

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1307. Min Act: -0.0001

Data Index: 1013938 (C4 (Web Text))

Max Activating Token Index: 58

Text #9

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1344. Min Act: -0.0001

Data Index: 266418 (C4 (Web Text))

Max Activating Token Index: 919

Text #10

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1296. Min Act: -0.0001

Data Index: 69194 (C4 (Web Text))

Max Activating Token Index: 97

Text #11

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1512. Min Act: -0.0001

Data Index: 439199 (C4 (Web Text))

Max Activating Token Index: 127

Text #12

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1281. Min Act: -0.0001

Data Index: 402760 (C4 (Web Text))

Max Activating Token Index: 522

Text #13

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1239. Min Act: -0.0001

Data Index: 692359 (C4 (Web Text))

Max Activating Token Index: 389

Text #14

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1293. Min Act: -0.0001

Data Index: 164368 (C4 (Web Text))

Max Activating Token Index: 489

Text #15

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1248. Min Act: -0.0001

Data Index: 81330 (C4 (Web Text))

Max Activating Token Index: 84

Text #16

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1288. Min Act: -0.0001

Data Index: 1610531 (Python Code)

Max Activating Token Index: 291

Text #17

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1195. Min Act: -0.0001

Data Index: 284914 (C4 (Web Text))

Max Activating Token Index: 851

Text #18

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1325. Min Act: -0.0001

Data Index: 438583 (C4 (Web Text))

Max Activating Token Index: 980

Text #19

Max Range: 0.1782. Min Range: -0.1782

Max Act: 0.1327. Min Act: -0.0001

Data Index: 63492 (C4 (Web Text))

Max Activating Token Index: 574