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

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

Neuron 3783 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.1180. Min Act: -0.0001

Data Index: 511771 (C4 (Web Text))

Max Activating Token Index: 317

Text #1

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.1107. Min Act: -0.0001

Data Index: 753166 (C4 (Web Text))

Max Activating Token Index: 615

Text #2

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.1088. Min Act: -0.0001

Data Index: 1479483 (Python Code)

Max Activating Token Index: 742

Text #3

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.1051. Min Act: -0.0001

Data Index: 487058 (C4 (Web Text))

Max Activating Token Index: 509

Text #4

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0946. Min Act: -0.0001

Data Index: 46675 (C4 (Web Text))

Max Activating Token Index: 676

Text #5

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0999. Min Act: -0.0001

Data Index: 1141464 (C4 (Web Text))

Max Activating Token Index: 360

Text #6

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0925. Min Act: -0.0001

Data Index: 69681 (C4 (Web Text))

Max Activating Token Index: 414

Text #7

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0886. Min Act: -0.0001

Data Index: 876073 (C4 (Web Text))

Max Activating Token Index: 673

Text #8

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0910. Min Act: -0.0001

Data Index: 948070 (C4 (Web Text))

Max Activating Token Index: 331

Text #9

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0983. Min Act: -0.0001

Data Index: 468770 (C4 (Web Text))

Max Activating Token Index: 724

Text #10

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0863. Min Act: -0.0001

Data Index: 299866 (C4 (Web Text))

Max Activating Token Index: 581

Text #11

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0888. Min Act: -0.0001

Data Index: 806724 (C4 (Web Text))

Max Activating Token Index: 533

Text #12

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0908. Min Act: -0.0001

Data Index: 262608 (C4 (Web Text))

Max Activating Token Index: 764

Text #13

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0780. Min Act: -0.0001

Data Index: 1620617 (Python Code)

Max Activating Token Index: 222

Text #14

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0866. Min Act: -0.0001

Data Index: 81581 (C4 (Web Text))

Max Activating Token Index: 455

Text #15

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0820. Min Act: -0.0001

Data Index: 995570 (C4 (Web Text))

Max Activating Token Index: 676

Text #16

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0763. Min Act: -0.0001

Data Index: 1256620 (C4 (Web Text))

Max Activating Token Index: 838

Text #17

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0850. Min Act: -0.0001

Data Index: 1330133 (C4 (Web Text))

Max Activating Token Index: 526

Text #18

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0906. Min Act: -0.0001

Data Index: 24252 (C4 (Web Text))

Max Activating Token Index: 319

Text #19

Max Range: 0.1180. Min Range: -0.1180

Max Act: 0.0781. Min Act: -0.0001

Data Index: 869145 (C4 (Web Text))

Max Activating Token Index: 835