Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3503 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.4147. Min Act: -0.0002

Data Index: 1231922 (C4 (Web Text))

Max Activating Token Index: 930

Text #1

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.4522. Min Act: -0.0002

Data Index: 1009859 (C4 (Web Text))

Max Activating Token Index: 12

Text #2

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3988. Min Act: -0.0002

Data Index: 1303649 (C4 (Web Text))

Max Activating Token Index: 893

Text #3

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.4276. Min Act: -0.0001

Data Index: 51658 (C4 (Web Text))

Max Activating Token Index: 312

Text #4

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.4020. Min Act: -0.0002

Data Index: 1217657 (C4 (Web Text))

Max Activating Token Index: 990

Text #5

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3835. Min Act: -0.0001

Data Index: 1463603 (Python Code)

Max Activating Token Index: 264

Text #6

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3711. Min Act: -0.0002

Data Index: 118313 (C4 (Web Text))

Max Activating Token Index: 76

Text #7

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3259. Min Act: -0.0001

Data Index: 1464527 (Python Code)

Max Activating Token Index: 694

Text #8

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3385. Min Act: -0.0002

Data Index: 749805 (C4 (Web Text))

Max Activating Token Index: 958

Text #9

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3514. Min Act: -0.0001

Data Index: 1058370 (C4 (Web Text))

Max Activating Token Index: 1020

Text #10

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3777. Min Act: -0.0002

Data Index: 353521 (C4 (Web Text))

Max Activating Token Index: 1012

Text #11

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3782. Min Act: -0.0002

Data Index: 353256 (C4 (Web Text))

Max Activating Token Index: 1013

Text #12

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3734. Min Act: -0.0001

Data Index: 1138838 (C4 (Web Text))

Max Activating Token Index: 782

Text #13

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3476. Min Act: -0.0001

Data Index: 378235 (C4 (Web Text))

Max Activating Token Index: 1009

Text #14

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3766. Min Act: -0.0002

Data Index: 159876 (C4 (Web Text))

Max Activating Token Index: 382

Text #15

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3578. Min Act: -0.0001

Data Index: 526172 (C4 (Web Text))

Max Activating Token Index: 948

Text #16

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3594. Min Act: -0.0002

Data Index: 1266650 (C4 (Web Text))

Max Activating Token Index: 1015

Text #17

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3560. Min Act: -0.0002

Data Index: 547379 (C4 (Web Text))

Max Activating Token Index: 977

Text #18

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.4222. Min Act: -0.0002

Data Index: 496649 (C4 (Web Text))

Max Activating Token Index: 1011

Text #19

Max Range: 0.4522. Min Range: -0.4522

Max Act: 0.3912. Min Act: -0.0002

Data Index: 36547 (C4 (Web Text))

Max Activating Token Index: 795