Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1646 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.8697. Min Act: -0.0005

Data Index: 96822 (C4 (Web Text))

Max Activating Token Index: 720

Text #1

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.8951. Min Act: -0.0005

Data Index: 875898 (C4 (Web Text))

Max Activating Token Index: 270

Text #2

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.7526. Min Act: -0.0005

Data Index: 1101539 (C4 (Web Text))

Max Activating Token Index: 752

Text #3

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.7041. Min Act: -0.0005

Data Index: 618698 (C4 (Web Text))

Max Activating Token Index: 333

Text #4

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.6783. Min Act: -0.0005

Data Index: 995374 (C4 (Web Text))

Max Activating Token Index: 414

Text #5

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.6484. Min Act: -0.0005

Data Index: 826568 (C4 (Web Text))

Max Activating Token Index: 637

Text #6

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.5855. Min Act: -0.0005

Data Index: 1293906 (C4 (Web Text))

Max Activating Token Index: 647

Text #7

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.5377. Min Act: -0.0005

Data Index: 892186 (C4 (Web Text))

Max Activating Token Index: 278

Text #8

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.6132. Min Act: -0.0005

Data Index: 222950 (C4 (Web Text))

Max Activating Token Index: 611

Text #9

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.6115. Min Act: -0.0005

Data Index: 45512 (C4 (Web Text))

Max Activating Token Index: 902

Text #10

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.5598. Min Act: -0.0005

Data Index: 156589 (C4 (Web Text))

Max Activating Token Index: 569

Text #11

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.5047. Min Act: -0.0005

Data Index: 236793 (C4 (Web Text))

Max Activating Token Index: 234

Text #12

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.4511. Min Act: -0.0005

Data Index: 752695 (C4 (Web Text))

Max Activating Token Index: 300

Text #13

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.5408. Min Act: -0.0005

Data Index: 1113029 (C4 (Web Text))

Max Activating Token Index: 282

Text #14

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.4053. Min Act: -0.0005

Data Index: 659880 (C4 (Web Text))

Max Activating Token Index: 475

Text #15

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.4041. Min Act: -0.0005

Data Index: 56685 (C4 (Web Text))

Max Activating Token Index: 785

Text #16

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.4852. Min Act: -0.0005

Data Index: 203389 (C4 (Web Text))

Max Activating Token Index: 178

Text #17

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.4765. Min Act: -0.0005

Data Index: 400503 (C4 (Web Text))

Max Activating Token Index: 619

Text #18

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.4455. Min Act: -0.0005

Data Index: 827963 (C4 (Web Text))

Max Activating Token Index: 460

Text #19

Max Range: 1.8951. Min Range: -1.8951

Max Act: 1.3512. Min Act: -0.0005

Data Index: 776640 (C4 (Web Text))

Max Activating Token Index: 366