Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4765 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.6967. Min Act: -0.0001

Data Index: 137932 (C4 (Web Text))

Max Activating Token Index: 212

Text #1

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.6971. Min Act: -0.0001

Data Index: 906486 (C4 (Web Text))

Max Activating Token Index: 595

Text #2

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.6838. Min Act: -0.0001

Data Index: 761802 (C4 (Web Text))

Max Activating Token Index: 21

Text #3

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.6785. Min Act: -0.0001

Data Index: 412647 (C4 (Web Text))

Max Activating Token Index: 358

Text #4

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.6367. Min Act: -0.0001

Data Index: 537385 (C4 (Web Text))

Max Activating Token Index: 740

Text #5

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.6791. Min Act: -0.0001

Data Index: 286531 (C4 (Web Text))

Max Activating Token Index: 1012

Text #6

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.6180. Min Act: -0.0001

Data Index: 261319 (C4 (Web Text))

Max Activating Token Index: 123

Text #7

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.6628. Min Act: -0.0001

Data Index: 1170177 (C4 (Web Text))

Max Activating Token Index: 346

Text #8

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.7044. Min Act: -0.0001

Data Index: 797962 (C4 (Web Text))

Max Activating Token Index: 26

Text #9

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.6692. Min Act: -0.0001

Data Index: 369941 (C4 (Web Text))

Max Activating Token Index: 661

Text #10

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.6559. Min Act: -0.0001

Data Index: 110712 (C4 (Web Text))

Max Activating Token Index: 432

Text #11

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.6065. Min Act: -0.0001

Data Index: 911060 (C4 (Web Text))

Max Activating Token Index: 938

Text #12

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.5765. Min Act: -0.0001

Data Index: 1005728 (C4 (Web Text))

Max Activating Token Index: 247

Text #13

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.5772. Min Act: -0.0001

Data Index: 858364 (C4 (Web Text))

Max Activating Token Index: 119

Text #14

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.5867. Min Act: -0.0001

Data Index: 946674 (C4 (Web Text))

Max Activating Token Index: 667

Text #15

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.5706. Min Act: -0.0001

Data Index: 1179037 (C4 (Web Text))

Max Activating Token Index: 62

Text #16

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.5450. Min Act: -0.0001

Data Index: 1234377 (C4 (Web Text))

Max Activating Token Index: 861

Text #17

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.5192. Min Act: -0.0001

Data Index: 593093 (C4 (Web Text))

Max Activating Token Index: 835

Text #18

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.6158. Min Act: -0.0001

Data Index: 58654 (C4 (Web Text))

Max Activating Token Index: 482

Text #19

Max Range: 0.7044. Min Range: -0.7044

Max Act: 0.5533. Min Act: -0.0001

Data Index: 986305 (C4 (Web Text))

Max Activating Token Index: 968