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

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

Neuron 1765 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.8168. Min Act: -0.0000

Data Index: 172676 (C4 (Web Text))

Max Activating Token Index: 853

Text #1

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.6500. Min Act: -0.0000

Data Index: 59773 (C4 (Web Text))

Max Activating Token Index: 504

Text #2

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.3404. Min Act: -0.0000

Data Index: 149860 (C4 (Web Text))

Max Activating Token Index: 314

Text #3

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.3776. Min Act: -0.0000

Data Index: 13959 (C4 (Web Text))

Max Activating Token Index: 704

Text #4

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.2593. Min Act: -0.0000

Data Index: 223309 (C4 (Web Text))

Max Activating Token Index: 571

Text #5

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.2227. Min Act: -0.0000

Data Index: 894899 (C4 (Web Text))

Max Activating Token Index: 436

Text #6

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.2582. Min Act: -0.0000

Data Index: 118345 (C4 (Web Text))

Max Activating Token Index: 84

Text #7

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.1510. Min Act: -0.0000

Data Index: 1291564 (C4 (Web Text))

Max Activating Token Index: 552

Text #8

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.0740. Min Act: -0.0000

Data Index: 945447 (C4 (Web Text))

Max Activating Token Index: 609

Text #9

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.1372. Min Act: -0.0000

Data Index: 470263 (C4 (Web Text))

Max Activating Token Index: 887

Text #10

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.1487. Min Act: -0.0000

Data Index: 54858 (C4 (Web Text))

Max Activating Token Index: 501

Text #11

Max Range: 1.8168. Min Range: -1.8168

Max Act: 0.9690. Min Act: -0.0000

Data Index: 667275 (C4 (Web Text))

Max Activating Token Index: 761

Text #12

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.0283. Min Act: -0.0000

Data Index: 1626258 (Python Code)

Max Activating Token Index: 809

Text #13

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.0889. Min Act: -0.0000

Data Index: 829952 (C4 (Web Text))

Max Activating Token Index: 1007

Text #14

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.1627. Min Act: -0.0000

Data Index: 252517 (C4 (Web Text))

Max Activating Token Index: 493

Text #15

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.0106. Min Act: -0.0000

Data Index: 149816 (C4 (Web Text))

Max Activating Token Index: 601

Text #16

Max Range: 1.8168. Min Range: -1.8168

Max Act: 1.1284. Min Act: -0.0000

Data Index: 1105070 (C4 (Web Text))

Max Activating Token Index: 723

Text #17

Max Range: 1.8168. Min Range: -1.8168

Max Act: 0.9554. Min Act: -0.0000

Data Index: 169512 (C4 (Web Text))

Max Activating Token Index: 326

Text #18

Max Range: 1.8168. Min Range: -1.8168

Max Act: 0.9226. Min Act: -0.0000

Data Index: 846920 (C4 (Web Text))

Max Activating Token Index: 526

Text #19

Max Range: 1.8168. Min Range: -1.8168

Max Act: 0.9641. Min Act: -0.0000

Data Index: 663866 (C4 (Web Text))

Max Activating Token Index: 1004