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

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

Neuron 1573 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.4136. Min Act: -0.0001

Data Index: 1593543 (Python Code)

Max Activating Token Index: 83

Text #1

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.3790. Min Act: -0.0001

Data Index: 888432 (C4 (Web Text))

Max Activating Token Index: 893

Text #2

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.3777. Min Act: -0.0001

Data Index: 1122563 (C4 (Web Text))

Max Activating Token Index: 262

Text #3

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.3513. Min Act: -0.0001

Data Index: 1137620 (C4 (Web Text))

Max Activating Token Index: 330

Text #4

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.3169. Min Act: -0.0001

Data Index: 1374598 (Python Code)

Max Activating Token Index: 847

Text #5

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.3150. Min Act: -0.0001

Data Index: 263167 (C4 (Web Text))

Max Activating Token Index: 859

Text #6

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.3092. Min Act: -0.0001

Data Index: 1067234 (C4 (Web Text))

Max Activating Token Index: 834

Text #7

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.3089. Min Act: -0.0001

Data Index: 1379108 (Python Code)

Max Activating Token Index: 46

Text #8

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.3088. Min Act: -0.0001

Data Index: 394693 (C4 (Web Text))

Max Activating Token Index: 535

Text #9

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.3021. Min Act: -0.0001

Data Index: 437942 (C4 (Web Text))

Max Activating Token Index: 784

Text #10

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.3109. Min Act: -0.0001

Data Index: 284490 (C4 (Web Text))

Max Activating Token Index: 580

Text #11

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.2772. Min Act: -0.0001

Data Index: 175973 (C4 (Web Text))

Max Activating Token Index: 861

Text #12

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.2669. Min Act: -0.0001

Data Index: 416308 (C4 (Web Text))

Max Activating Token Index: 999

Text #13

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.2681. Min Act: -0.0001

Data Index: 1558227 (Python Code)

Max Activating Token Index: 22

Text #14

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.2469. Min Act: -0.0001

Data Index: 1365382 (Python Code)

Max Activating Token Index: 657

Text #15

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.2666. Min Act: -0.0001

Data Index: 1648016 (Python Code)

Max Activating Token Index: 72

Text #16

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.2685. Min Act: -0.0001

Data Index: 1228395 (C4 (Web Text))

Max Activating Token Index: 675

Text #17

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.2414. Min Act: -0.0001

Data Index: 49538 (C4 (Web Text))

Max Activating Token Index: 992

Text #18

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.2662. Min Act: -0.0001

Data Index: 24361 (C4 (Web Text))

Max Activating Token Index: 173

Text #19

Max Range: 0.4136. Min Range: -0.4136

Max Act: 0.2618. Min Act: -0.0001

Data Index: 228505 (C4 (Web Text))

Max Activating Token Index: 548