Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 2680 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 4.5934. Min Range: -4.5934

Max Act: 4.5934. Min Act: -0.0000

Data Index: 1369978 (Python Code)

Max Activating Token Index: 890

Text #1

Max Range: 4.5934. Min Range: -4.5934

Max Act: 4.4305. Min Act: -0.0000

Data Index: 995609 (C4 (Web Text))

Max Activating Token Index: 375

Text #2

Max Range: 4.5934. Min Range: -4.5934

Max Act: 4.4031. Min Act: -0.0000

Data Index: 368212 (C4 (Web Text))

Max Activating Token Index: 355

Text #3

Max Range: 4.5934. Min Range: -4.5934

Max Act: 4.1840. Min Act: -0.0000

Data Index: 1074115 (C4 (Web Text))

Max Activating Token Index: 923

Text #4

Max Range: 4.5934. Min Range: -4.5934

Max Act: 4.1975. Min Act: -0.0000

Data Index: 568306 (C4 (Web Text))

Max Activating Token Index: 554

Text #5

Max Range: 4.5934. Min Range: -4.5934

Max Act: 4.1814. Min Act: -0.0000

Data Index: 835935 (C4 (Web Text))

Max Activating Token Index: 283

Text #6

Max Range: 4.5934. Min Range: -4.5934

Max Act: 3.8870. Min Act: -0.0000

Data Index: 888269 (C4 (Web Text))

Max Activating Token Index: 469

Text #7

Max Range: 4.5934. Min Range: -4.5934

Max Act: 4.1986. Min Act: -0.0000

Data Index: 1044787 (C4 (Web Text))

Max Activating Token Index: 785

Text #8

Max Range: 4.5934. Min Range: -4.5934

Max Act: 4.0342. Min Act: -0.0000

Data Index: 1195091 (C4 (Web Text))

Max Activating Token Index: 296

Text #9

Max Range: 4.5934. Min Range: -4.5934

Max Act: 3.6756. Min Act: -0.0000

Data Index: 99735 (C4 (Web Text))

Max Activating Token Index: 419

Text #10

Max Range: 4.5934. Min Range: -4.5934

Max Act: 3.9196. Min Act: -0.0000

Data Index: 1550617 (Python Code)

Max Activating Token Index: 912

Text #11

Max Range: 4.5934. Min Range: -4.5934

Max Act: 4.1198. Min Act: -0.0000

Data Index: 264784 (C4 (Web Text))

Max Activating Token Index: 799

Text #12

Max Range: 4.5934. Min Range: -4.5934

Max Act: 4.0867. Min Act: -0.0000

Data Index: 1624418 (Python Code)

Max Activating Token Index: 302

Text #13

Max Range: 4.5934. Min Range: -4.5934

Max Act: 3.7323. Min Act: -0.0000

Data Index: 471244 (C4 (Web Text))

Max Activating Token Index: 323

Text #14

Max Range: 4.5934. Min Range: -4.5934

Max Act: 3.5336. Min Act: -0.0000

Data Index: 802819 (C4 (Web Text))

Max Activating Token Index: 443

Text #15

Max Range: 4.5934. Min Range: -4.5934

Max Act: 3.4836. Min Act: -0.0000

Data Index: 334815 (C4 (Web Text))

Max Activating Token Index: 391

Text #16

Max Range: 4.5934. Min Range: -4.5934

Max Act: 3.6362. Min Act: -0.0000

Data Index: 1652690 (Python Code)

Max Activating Token Index: 703

Text #17

Max Range: 4.5934. Min Range: -4.5934

Max Act: 3.6103. Min Act: -0.0000

Data Index: 1249586 (C4 (Web Text))

Max Activating Token Index: 115

Text #18

Max Range: 4.5934. Min Range: -4.5934

Max Act: 3.7539. Min Act: -0.0000

Data Index: 1305785 (C4 (Web Text))

Max Activating Token Index: 640

Text #19

Max Range: 4.5934. Min Range: -4.5934

Max Act: 3.3832. Min Act: -0.0000

Data Index: 1579499 (Python Code)

Max Activating Token Index: 649