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

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

Neuron 33 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: 0.2513. Min Range: -0.2513

Max Act: 0.2493. Min Act: -0.0005

Data Index: 903205 (C4 (Web Text))

Max Activating Token Index: 954

Text #1

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2363. Min Act: -0.0005

Data Index: 1246704 (C4 (Web Text))

Max Activating Token Index: 1015

Text #2

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2356. Min Act: -0.0005

Data Index: 8310 (C4 (Web Text))

Max Activating Token Index: 881

Text #3

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2513. Min Act: -0.0005

Data Index: 124813 (C4 (Web Text))

Max Activating Token Index: 277

Text #4

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2403. Min Act: -0.0005

Data Index: 665874 (C4 (Web Text))

Max Activating Token Index: 739

Text #5

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2322. Min Act: -0.0005

Data Index: 409028 (C4 (Web Text))

Max Activating Token Index: 97

Text #6

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2282. Min Act: -0.0005

Data Index: 134241 (C4 (Web Text))

Max Activating Token Index: 929

Text #7

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2230. Min Act: -0.0005

Data Index: 830403 (C4 (Web Text))

Max Activating Token Index: 132

Text #8

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2329. Min Act: -0.0005

Data Index: 4028 (C4 (Web Text))

Max Activating Token Index: 770

Text #9

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2142. Min Act: -0.0005

Data Index: 55769 (C4 (Web Text))

Max Activating Token Index: 883

Text #10

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2268. Min Act: -0.0005

Data Index: 254553 (C4 (Web Text))

Max Activating Token Index: 882

Text #11

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2162. Min Act: -0.0005

Data Index: 686441 (C4 (Web Text))

Max Activating Token Index: 568

Text #12

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.1978. Min Act: -0.0005

Data Index: 1026027 (C4 (Web Text))

Max Activating Token Index: 793

Text #13

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2101. Min Act: -0.0005

Data Index: 891251 (C4 (Web Text))

Max Activating Token Index: 94

Text #14

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2041. Min Act: -0.0005

Data Index: 1070637 (C4 (Web Text))

Max Activating Token Index: 863

Text #15

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2023. Min Act: -0.0005

Data Index: 770415 (C4 (Web Text))

Max Activating Token Index: 780

Text #16

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2155. Min Act: -0.0005

Data Index: 689003 (C4 (Web Text))

Max Activating Token Index: 842

Text #17

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.2171. Min Act: -0.0005

Data Index: 229359 (C4 (Web Text))

Max Activating Token Index: 613

Text #18

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.1998. Min Act: -0.0005

Data Index: 1076323 (C4 (Web Text))

Max Activating Token Index: 362

Text #19

Max Range: 0.2513. Min Range: -0.2513

Max Act: 0.1986. Min Act: -0.0005

Data Index: 374954 (C4 (Web Text))

Max Activating Token Index: 352