Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 2146 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3755. Min Act: -0.0000

Data Index: 1549528 (Python Code)

Max Activating Token Index: 674

Text #1

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3501. Min Act: -0.0000

Data Index: 976688 (C4 (Web Text))

Max Activating Token Index: 721

Text #2

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3960. Min Act: -0.0000

Data Index: 1011262 (C4 (Web Text))

Max Activating Token Index: 113

Text #3

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3822. Min Act: -0.0000

Data Index: 968570 (C4 (Web Text))

Max Activating Token Index: 766

Text #4

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3588. Min Act: -0.0000

Data Index: 1213081 (C4 (Web Text))

Max Activating Token Index: 929

Text #5

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3414. Min Act: -0.0000

Data Index: 302174 (C4 (Web Text))

Max Activating Token Index: 481

Text #6

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3441. Min Act: -0.0000

Data Index: 1232554 (C4 (Web Text))

Max Activating Token Index: 855

Text #7

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3530. Min Act: -0.0000

Data Index: 553897 (C4 (Web Text))

Max Activating Token Index: 667

Text #8

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3257. Min Act: -0.0000

Data Index: 1410298 (Python Code)

Max Activating Token Index: 812

Text #9

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3360. Min Act: -0.0000

Data Index: 1518423 (Python Code)

Max Activating Token Index: 462

Text #10

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3270. Min Act: -0.0000

Data Index: 102053 (C4 (Web Text))

Max Activating Token Index: 327

Text #11

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3196. Min Act: -0.0000

Data Index: 1211684 (C4 (Web Text))

Max Activating Token Index: 439

Text #12

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3190. Min Act: -0.0000

Data Index: 1576887 (Python Code)

Max Activating Token Index: 1014

Text #13

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3307. Min Act: -0.0000

Data Index: 290828 (C4 (Web Text))

Max Activating Token Index: 1011

Text #14

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3076. Min Act: -0.0000

Data Index: 930220 (C4 (Web Text))

Max Activating Token Index: 856

Text #15

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.2925. Min Act: -0.0000

Data Index: 1323037 (C4 (Web Text))

Max Activating Token Index: 613

Text #16

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.3252. Min Act: -0.0000

Data Index: 1587924 (Python Code)

Max Activating Token Index: 536

Text #17

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.2861. Min Act: -0.0000

Data Index: 1652339 (Python Code)

Max Activating Token Index: 950

Text #18

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.2967. Min Act: -0.0000

Data Index: 606424 (C4 (Web Text))

Max Activating Token Index: 889

Text #19

Max Range: 0.3960. Min Range: -0.3960

Max Act: 0.2942. Min Act: -0.0000

Data Index: 615352 (C4 (Web Text))

Max Activating Token Index: 1013