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

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

Neuron 3146 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: 0.8950. Min Range: -0.8950

Max Act: 0.8950. Min Act: -0.0002

Data Index: 162661 (C4 (Web Text))

Max Activating Token Index: 43

Text #1

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.8476. Min Act: -0.0001

Data Index: 807798 (C4 (Web Text))

Max Activating Token Index: 434

Text #2

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.8416. Min Act: -0.0001

Data Index: 858569 (C4 (Web Text))

Max Activating Token Index: 810

Text #3

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.8750. Min Act: -0.0001

Data Index: 783840 (C4 (Web Text))

Max Activating Token Index: 226

Text #4

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.8133. Min Act: -0.0001

Data Index: 884760 (C4 (Web Text))

Max Activating Token Index: 27

Text #5

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.7840. Min Act: -0.0001

Data Index: 879803 (C4 (Web Text))

Max Activating Token Index: 606

Text #6

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.7543. Min Act: -0.0001

Data Index: 651628 (C4 (Web Text))

Max Activating Token Index: 445

Text #7

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.7659. Min Act: -0.0002

Data Index: 1167399 (C4 (Web Text))

Max Activating Token Index: 43

Text #8

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.7769. Min Act: -0.0001

Data Index: 929115 (C4 (Web Text))

Max Activating Token Index: 655

Text #9

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.7398. Min Act: -0.0001

Data Index: 197269 (C4 (Web Text))

Max Activating Token Index: 38

Text #10

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.6756. Min Act: -0.0002

Data Index: 228245 (C4 (Web Text))

Max Activating Token Index: 768

Text #11

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.7030. Min Act: -0.0001

Data Index: 1408818 (Python Code)

Max Activating Token Index: 913

Text #12

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.6826. Min Act: -0.0001

Data Index: 1564723 (Python Code)

Max Activating Token Index: 1008

Text #13

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.7322. Min Act: -0.0001

Data Index: 1566204 (Python Code)

Max Activating Token Index: 891

Text #14

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.7305. Min Act: -0.0002

Data Index: 308532 (C4 (Web Text))

Max Activating Token Index: 56

Text #15

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.7244. Min Act: -0.0002

Data Index: 562665 (C4 (Web Text))

Max Activating Token Index: 863

Text #16

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.6356. Min Act: -0.0002

Data Index: 627179 (C4 (Web Text))

Max Activating Token Index: 774

Text #17

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.7019. Min Act: -0.0002

Data Index: 360935 (C4 (Web Text))

Max Activating Token Index: 81

Text #18

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.6416. Min Act: -0.0001

Data Index: 1371368 (Python Code)

Max Activating Token Index: 880

Text #19

Max Range: 0.8950. Min Range: -0.8950

Max Act: 0.6903. Min Act: -0.0002

Data Index: 1318559 (C4 (Web Text))

Max Activating Token Index: 52