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

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

Neuron 4871 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: 1.4862. Min Range: -1.4862

Max Act: 1.4862. Min Act: -0.0001

Data Index: 1294263 (C4 (Web Text))

Max Activating Token Index: 910

Text #1

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.4763. Min Act: -0.0001

Data Index: 348626 (C4 (Web Text))

Max Activating Token Index: 471

Text #2

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.4182. Min Act: -0.0001

Data Index: 333448 (C4 (Web Text))

Max Activating Token Index: 27

Text #3

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.3605. Min Act: -0.0001

Data Index: 1255428 (C4 (Web Text))

Max Activating Token Index: 443

Text #4

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.4048. Min Act: -0.0001

Data Index: 57000 (C4 (Web Text))

Max Activating Token Index: 797

Text #5

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.3441. Min Act: -0.0001

Data Index: 1294508 (C4 (Web Text))

Max Activating Token Index: 377

Text #6

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.2450. Min Act: -0.0001

Data Index: 860625 (C4 (Web Text))

Max Activating Token Index: 244

Text #7

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.2603. Min Act: -0.0001

Data Index: 112390 (C4 (Web Text))

Max Activating Token Index: 591

Text #8

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.3147. Min Act: -0.0001

Data Index: 153144 (C4 (Web Text))

Max Activating Token Index: 371

Text #9

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.3147. Min Act: -0.0001

Data Index: 365299 (C4 (Web Text))

Max Activating Token Index: 722

Text #10

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.3317. Min Act: -0.0001

Data Index: 524074 (C4 (Web Text))

Max Activating Token Index: 672

Text #11

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.2649. Min Act: -0.0001

Data Index: 270434 (C4 (Web Text))

Max Activating Token Index: 891

Text #12

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.2220. Min Act: -0.0001

Data Index: 412306 (C4 (Web Text))

Max Activating Token Index: 397

Text #13

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.2553. Min Act: -0.0001

Data Index: 1252808 (C4 (Web Text))

Max Activating Token Index: 653

Text #14

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.2875. Min Act: -0.0001

Data Index: 73023 (C4 (Web Text))

Max Activating Token Index: 5

Text #15

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.2496. Min Act: -0.0001

Data Index: 353008 (C4 (Web Text))

Max Activating Token Index: 224

Text #16

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.2779. Min Act: -0.0001

Data Index: 580190 (C4 (Web Text))

Max Activating Token Index: 553

Text #17

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.2772. Min Act: -0.0001

Data Index: 539889 (C4 (Web Text))

Max Activating Token Index: 749

Text #18

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.2588. Min Act: -0.0001

Data Index: 1311684 (C4 (Web Text))

Max Activating Token Index: 267

Text #19

Max Range: 1.4862. Min Range: -1.4862

Max Act: 1.2860. Min Act: -0.0001

Data Index: 695666 (C4 (Web Text))

Max Activating Token Index: 138