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

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

Neuron 2921 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.9576. Min Act: -0.0001

Data Index: 1647524 (Python Code)

Max Activating Token Index: 549

Text #1

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.7641. Min Act: -0.0001

Data Index: 1469592 (Python Code)

Max Activating Token Index: 810

Text #2

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.8246. Min Act: -0.0001

Data Index: 1568949 (Python Code)

Max Activating Token Index: 952

Text #3

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.8384. Min Act: -0.0001

Data Index: 1071494 (C4 (Web Text))

Max Activating Token Index: 11

Text #4

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.7558. Min Act: -0.0001

Data Index: 1639361 (Python Code)

Max Activating Token Index: 594

Text #5

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.7337. Min Act: -0.0001

Data Index: 205638 (C4 (Web Text))

Max Activating Token Index: 726

Text #6

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.7395. Min Act: -0.0001

Data Index: 21718 (C4 (Web Text))

Max Activating Token Index: 992

Text #7

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.7188. Min Act: -0.0001

Data Index: 1634676 (Python Code)

Max Activating Token Index: 493

Text #8

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.7565. Min Act: -0.0001

Data Index: 1193672 (C4 (Web Text))

Max Activating Token Index: 596

Text #9

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.6960. Min Act: -0.0001

Data Index: 1449888 (Python Code)

Max Activating Token Index: 705

Text #10

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.6593. Min Act: -0.0001

Data Index: 215352 (C4 (Web Text))

Max Activating Token Index: 467

Text #11

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.6737. Min Act: -0.0001

Data Index: 1053457 (C4 (Web Text))

Max Activating Token Index: 579

Text #12

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.6774. Min Act: -0.0001

Data Index: 169043 (C4 (Web Text))

Max Activating Token Index: 322

Text #13

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.6393. Min Act: -0.0001

Data Index: 875697 (C4 (Web Text))

Max Activating Token Index: 443

Text #14

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.6964. Min Act: -0.0001

Data Index: 466967 (C4 (Web Text))

Max Activating Token Index: 97

Text #15

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.6451. Min Act: -0.0001

Data Index: 710976 (C4 (Web Text))

Max Activating Token Index: 27

Text #16

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.6832. Min Act: -0.0001

Data Index: 771468 (C4 (Web Text))

Max Activating Token Index: 74

Text #17

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.7094. Min Act: -0.0001

Data Index: 1299390 (C4 (Web Text))

Max Activating Token Index: 806

Text #18

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.6598. Min Act: -0.0001

Data Index: 798107 (C4 (Web Text))

Max Activating Token Index: 530

Text #19

Max Range: 0.9576. Min Range: -0.9576

Max Act: 0.6785. Min Act: -0.0001

Data Index: 333308 (C4 (Web Text))

Max Activating Token Index: 522