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

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

Neuron 3647 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.6476. Min Act: -0.0001

Data Index: 1491966 (Python Code)

Max Activating Token Index: 557

Text #1

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.7414. Min Act: -0.0001

Data Index: 1635098 (Python Code)

Max Activating Token Index: 563

Text #2

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.4775. Min Act: -0.0001

Data Index: 873646 (C4 (Web Text))

Max Activating Token Index: 632

Text #3

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.4433. Min Act: -0.0001

Data Index: 1434638 (Python Code)

Max Activating Token Index: 462

Text #4

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.4734. Min Act: -0.0001

Data Index: 640901 (C4 (Web Text))

Max Activating Token Index: 852

Text #5

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.3866. Min Act: -0.0001

Data Index: 793688 (C4 (Web Text))

Max Activating Token Index: 772

Text #6

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.3824. Min Act: -0.0001

Data Index: 1525105 (Python Code)

Max Activating Token Index: 631

Text #7

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.4573. Min Act: -0.0001

Data Index: 1616273 (Python Code)

Max Activating Token Index: 537

Text #8

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.4343. Min Act: -0.0001

Data Index: 730849 (C4 (Web Text))

Max Activating Token Index: 499

Text #9

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.3942. Min Act: -0.0001

Data Index: 1412318 (Python Code)

Max Activating Token Index: 105

Text #10

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.3055. Min Act: -0.0001

Data Index: 83907 (C4 (Web Text))

Max Activating Token Index: 537

Text #11

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.3705. Min Act: -0.0001

Data Index: 75605 (C4 (Web Text))

Max Activating Token Index: 673

Text #12

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.4105. Min Act: -0.0001

Data Index: 1558187 (Python Code)

Max Activating Token Index: 322

Text #13

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.3536. Min Act: -0.0001

Data Index: 1655263 (Python Code)

Max Activating Token Index: 887

Text #14

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.3413. Min Act: -0.0001

Data Index: 1584116 (Python Code)

Max Activating Token Index: 279

Text #15

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.3847. Min Act: -0.0001

Data Index: 1498825 (Python Code)

Max Activating Token Index: 469

Text #16

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.4142. Min Act: -0.0001

Data Index: 1453963 (Python Code)

Max Activating Token Index: 680

Text #17

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.3239. Min Act: -0.0001

Data Index: 371289 (C4 (Web Text))

Max Activating Token Index: 857

Text #18

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.3134. Min Act: -0.0001

Data Index: 1446404 (Python Code)

Max Activating Token Index: 475

Text #19

Max Range: 1.7414. Min Range: -1.7414

Max Act: 1.3016. Min Act: -0.0001

Data Index: 1470594 (Python Code)

Max Activating Token Index: 913