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

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

Neuron 2882 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.2338. Min Range: -1.2338

Max Act: 1.2338. Min Act: -0.0001

Data Index: 361156 (C4 (Web Text))

Max Activating Token Index: 950

Text #1

Max Range: 1.2338. Min Range: -1.2338

Max Act: 1.2222. Min Act: -0.0001

Data Index: 1217501 (C4 (Web Text))

Max Activating Token Index: 912

Text #2

Max Range: 1.2338. Min Range: -1.2338

Max Act: 1.1470. Min Act: -0.0001

Data Index: 145082 (C4 (Web Text))

Max Activating Token Index: 208

Text #3

Max Range: 1.2338. Min Range: -1.2338

Max Act: 1.0622. Min Act: -0.0001

Data Index: 726412 (C4 (Web Text))

Max Activating Token Index: 447

Text #4

Max Range: 1.2338. Min Range: -1.2338

Max Act: 1.0217. Min Act: -0.0001

Data Index: 1195419 (C4 (Web Text))

Max Activating Token Index: 34

Text #5

Max Range: 1.2338. Min Range: -1.2338

Max Act: 1.0383. Min Act: -0.0001

Data Index: 1474627 (Python Code)

Max Activating Token Index: 812

Text #6

Max Range: 1.2338. Min Range: -1.2338

Max Act: 0.9895. Min Act: -0.0001

Data Index: 377311 (C4 (Web Text))

Max Activating Token Index: 235

Text #7

Max Range: 1.2338. Min Range: -1.2338

Max Act: 0.9848. Min Act: -0.0001

Data Index: 1522737 (Python Code)

Max Activating Token Index: 784

Text #8

Max Range: 1.2338. Min Range: -1.2338

Max Act: 0.9745. Min Act: -0.0000

Data Index: 480769 (C4 (Web Text))

Max Activating Token Index: 253

Text #9

Max Range: 1.2338. Min Range: -1.2338

Max Act: 0.9756. Min Act: -0.0001

Data Index: 1405926 (Python Code)

Max Activating Token Index: 881

Text #10

Max Range: 1.2338. Min Range: -1.2338

Max Act: 0.9792. Min Act: -0.0001

Data Index: 378828 (C4 (Web Text))

Max Activating Token Index: 25

Text #11

Max Range: 1.2338. Min Range: -1.2338

Max Act: 1.0351. Min Act: -0.0000

Data Index: 334371 (C4 (Web Text))

Max Activating Token Index: 954

Text #12

Max Range: 1.2338. Min Range: -1.2338

Max Act: 1.0694. Min Act: -0.0001

Data Index: 553465 (C4 (Web Text))

Max Activating Token Index: 518

Text #13

Max Range: 1.2338. Min Range: -1.2338

Max Act: 0.9953. Min Act: -0.0000

Data Index: 140313 (C4 (Web Text))

Max Activating Token Index: 515

Text #14

Max Range: 1.2338. Min Range: -1.2338

Max Act: 0.9479. Min Act: -0.0001

Data Index: 1381956 (Python Code)

Max Activating Token Index: 988

Text #15

Max Range: 1.2338. Min Range: -1.2338

Max Act: 0.8778. Min Act: -0.0001

Data Index: 1320234 (C4 (Web Text))

Max Activating Token Index: 797

Text #16

Max Range: 1.2338. Min Range: -1.2338

Max Act: 0.9999. Min Act: -0.0001

Data Index: 1604515 (Python Code)

Max Activating Token Index: 907

Text #17

Max Range: 1.2338. Min Range: -1.2338

Max Act: 0.9624. Min Act: -0.0001

Data Index: 1462219 (Python Code)

Max Activating Token Index: 521

Text #18

Max Range: 1.2338. Min Range: -1.2338

Max Act: 0.9369. Min Act: -0.0001

Data Index: 1211248 (C4 (Web Text))

Max Activating Token Index: 838

Text #19

Max Range: 1.2338. Min Range: -1.2338

Max Act: 0.9366. Min Act: -0.0001

Data Index: 1031195 (C4 (Web Text))

Max Activating Token Index: 828