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

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

Neuron 2835 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1422. Min Range: -3.1422

Max Act: 3.0061. Min Act: -0.0001

Data Index: 1250619 (C4 (Web Text))

Max Activating Token Index: 65

Text #1

Max Range: 3.1422. Min Range: -3.1422

Max Act: 3.1422. Min Act: -0.0001

Data Index: 849258 (C4 (Web Text))

Max Activating Token Index: 35

Text #2

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.6999. Min Act: -0.0001

Data Index: 1103971 (C4 (Web Text))

Max Activating Token Index: 199

Text #3

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.7616. Min Act: -0.0001

Data Index: 619627 (C4 (Web Text))

Max Activating Token Index: 937

Text #4

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.7281. Min Act: -0.0001

Data Index: 783313 (C4 (Web Text))

Max Activating Token Index: 822

Text #5

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.6829. Min Act: -0.0001

Data Index: 756282 (C4 (Web Text))

Max Activating Token Index: 21

Text #6

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.6104. Min Act: -0.0001

Data Index: 619068 (C4 (Web Text))

Max Activating Token Index: 41

Text #7

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.5907. Min Act: -0.0001

Data Index: 1025514 (C4 (Web Text))

Max Activating Token Index: 80

Text #8

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.5908. Min Act: -0.0001

Data Index: 967773 (C4 (Web Text))

Max Activating Token Index: 25

Text #9

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.5848. Min Act: -0.0001

Data Index: 82821 (C4 (Web Text))

Max Activating Token Index: 14

Text #10

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.4712. Min Act: -0.0001

Data Index: 484219 (C4 (Web Text))

Max Activating Token Index: 848

Text #11

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.5085. Min Act: -0.0001

Data Index: 1064155 (C4 (Web Text))

Max Activating Token Index: 455

Text #12

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.4860. Min Act: -0.0001

Data Index: 1032405 (C4 (Web Text))

Max Activating Token Index: 599

Text #13

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.4287. Min Act: -0.0001

Data Index: 663362 (C4 (Web Text))

Max Activating Token Index: 260

Text #14

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.4149. Min Act: -0.0001

Data Index: 211936 (C4 (Web Text))

Max Activating Token Index: 32

Text #15

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.3991. Min Act: -0.0001

Data Index: 457885 (C4 (Web Text))

Max Activating Token Index: 879

Text #16

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.3982. Min Act: -0.0001

Data Index: 785966 (C4 (Web Text))

Max Activating Token Index: 946

Text #17

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.4031. Min Act: -0.0001

Data Index: 369970 (C4 (Web Text))

Max Activating Token Index: 978

Text #18

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.3188. Min Act: -0.0001

Data Index: 1147946 (C4 (Web Text))

Max Activating Token Index: 336

Text #19

Max Range: 3.1422. Min Range: -3.1422

Max Act: 2.3417. Min Act: -0.0001

Data Index: 1173100 (C4 (Web Text))

Max Activating Token Index: 14