Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5543 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 15.8610. Min Range: -15.8610

Max Act: 15.8610. Min Act: -0.0000

Data Index: 862674 (C4 (Web Text))

Max Activating Token Index: 326

Text #1

Max Range: 15.8610. Min Range: -15.8610

Max Act: 13.5628. Min Act: -0.0000

Data Index: 561392 (C4 (Web Text))

Max Activating Token Index: 163

Text #2

Max Range: 15.8610. Min Range: -15.8610

Max Act: 12.2982. Min Act: -0.0000

Data Index: 858227 (C4 (Web Text))

Max Activating Token Index: 41

Text #3

Max Range: 15.8610. Min Range: -15.8610

Max Act: 0.0261. Min Act: -0.0000

Data Index: 1272840 (C4 (Web Text))

Max Activating Token Index: 448

Text #4

Max Range: 15.8610. Min Range: -15.8610

Max Act: 3.8634. Min Act: -0.0000

Data Index: 367983 (C4 (Web Text))

Max Activating Token Index: 1022

Text #5

Max Range: 15.8610. Min Range: -15.8610

Max Act: 11.1219. Min Act: -0.0000

Data Index: 913268 (C4 (Web Text))

Max Activating Token Index: 966

Text #6

Max Range: 15.8610. Min Range: -15.8610

Max Act: 7.7352. Min Act: -0.0000

Data Index: 142020 (C4 (Web Text))

Max Activating Token Index: 951

Text #7

Max Range: 15.8610. Min Range: -15.8610

Max Act: 2.3061. Min Act: -0.0000

Data Index: 1115708 (C4 (Web Text))

Max Activating Token Index: 294

Text #8

Max Range: 15.8610. Min Range: -15.8610

Max Act: 7.9771. Min Act: -0.0000

Data Index: 589438 (C4 (Web Text))

Max Activating Token Index: 590

Text #9

Max Range: 15.8610. Min Range: -15.8610

Max Act: 8.4775. Min Act: -0.0000

Data Index: 547199 (C4 (Web Text))

Max Activating Token Index: 941

Text #10

Max Range: 15.8610. Min Range: -15.8610

Max Act: 4.1307. Min Act: -0.0000

Data Index: 918918 (C4 (Web Text))

Max Activating Token Index: 657

Text #11

Max Range: 15.8610. Min Range: -15.8610

Max Act: 11.0700. Min Act: -0.0000

Data Index: 509853 (C4 (Web Text))

Max Activating Token Index: 761

Text #12

Max Range: 15.8610. Min Range: -15.8610

Max Act: 5.1104. Min Act: -0.0000

Data Index: 766278 (C4 (Web Text))

Max Activating Token Index: 790

Text #13

Max Range: 15.8610. Min Range: -15.8610

Max Act: 5.3264. Min Act: -0.0000

Data Index: 131973 (C4 (Web Text))

Max Activating Token Index: 285

Text #14

Max Range: 15.8610. Min Range: -15.8610

Max Act: 3.2633. Min Act: -0.0000

Data Index: 1638559 (Python Code)

Max Activating Token Index: 60

Text #15

Max Range: 15.8610. Min Range: -15.8610

Max Act: 2.6843. Min Act: -0.0000

Data Index: 692258 (C4 (Web Text))

Max Activating Token Index: 659

Text #16

Max Range: 15.8610. Min Range: -15.8610

Max Act: 2.7297. Min Act: -0.0000

Data Index: 981748 (C4 (Web Text))

Max Activating Token Index: 609

Text #17

Max Range: 15.8610. Min Range: -15.8610

Max Act: 11.4585. Min Act: -0.0000

Data Index: 1560375 (Python Code)

Max Activating Token Index: 317

Text #18

Max Range: 15.8610. Min Range: -15.8610

Max Act: 2.0131. Min Act: -0.0000

Data Index: 1653736 (Python Code)

Max Activating Token Index: 241

Text #19

Max Range: 15.8610. Min Range: -15.8610

Max Act: 0.0167. Min Act: -0.0000

Data Index: 403981 (C4 (Web Text))

Max Activating Token Index: 967