Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 796 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0468. Min Act: -0.0001

Data Index: 666773 (C4 (Web Text))

Max Activating Token Index: 696

Text #1

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0483. Min Act: -0.0001

Data Index: 518294 (C4 (Web Text))

Max Activating Token Index: 242

Text #2

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0437. Min Act: -0.0001

Data Index: 31062 (C4 (Web Text))

Max Activating Token Index: 13

Text #3

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0406. Min Act: -0.0001

Data Index: 294589 (C4 (Web Text))

Max Activating Token Index: 949

Text #4

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0409. Min Act: -0.0001

Data Index: 937463 (C4 (Web Text))

Max Activating Token Index: 978

Text #5

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0393. Min Act: -0.0001

Data Index: 822851 (C4 (Web Text))

Max Activating Token Index: 511

Text #6

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0407. Min Act: -0.0001

Data Index: 1071581 (C4 (Web Text))

Max Activating Token Index: 885

Text #7

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0390. Min Act: -0.0001

Data Index: 862771 (C4 (Web Text))

Max Activating Token Index: 389

Text #8

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0379. Min Act: -0.0001

Data Index: 788455 (C4 (Web Text))

Max Activating Token Index: 62

Text #9

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0366. Min Act: -0.0001

Data Index: 778621 (C4 (Web Text))

Max Activating Token Index: 680

Text #10

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0364. Min Act: -0.0001

Data Index: 634518 (C4 (Web Text))

Max Activating Token Index: 968

Text #11

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0369. Min Act: -0.0001

Data Index: 594636 (C4 (Web Text))

Max Activating Token Index: 947

Text #12

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0375. Min Act: -0.0001

Data Index: 245801 (C4 (Web Text))

Max Activating Token Index: 963

Text #13

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0347. Min Act: -0.0001

Data Index: 173278 (C4 (Web Text))

Max Activating Token Index: 705

Text #14

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0358. Min Act: -0.0001

Data Index: 627223 (C4 (Web Text))

Max Activating Token Index: 228

Text #15

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0353. Min Act: -0.0001

Data Index: 867152 (C4 (Web Text))

Max Activating Token Index: 134

Text #16

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0352. Min Act: -0.0001

Data Index: 727495 (C4 (Web Text))

Max Activating Token Index: 1012

Text #17

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0356. Min Act: -0.0001

Data Index: 1472553 (Python Code)

Max Activating Token Index: 570

Text #18

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0352. Min Act: -0.0001

Data Index: 974652 (C4 (Web Text))

Max Activating Token Index: 910

Text #19

Max Range: 0.0483. Min Range: -0.0483

Max Act: 0.0353. Min Act: -0.0001

Data Index: 1143635 (C4 (Web Text))

Max Activating Token Index: 151