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

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

Neuron 3421 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: 0.1266. Min Range: -0.1266

Max Act: 0.1253. Min Act: -0.0000

Data Index: 988880 (C4 (Web Text))

Max Activating Token Index: 779

Text #1

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.1266. Min Act: -0.0000

Data Index: 747223 (C4 (Web Text))

Max Activating Token Index: 818

Text #2

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.1018. Min Act: -0.0000

Data Index: 187330 (C4 (Web Text))

Max Activating Token Index: 756

Text #3

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.1135. Min Act: -0.0000

Data Index: 704418 (C4 (Web Text))

Max Activating Token Index: 713

Text #4

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.1042. Min Act: -0.0001

Data Index: 242695 (C4 (Web Text))

Max Activating Token Index: 384

Text #5

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.1034. Min Act: -0.0000

Data Index: 80030 (C4 (Web Text))

Max Activating Token Index: 818

Text #6

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.1044. Min Act: -0.0000

Data Index: 1163157 (C4 (Web Text))

Max Activating Token Index: 966

Text #7

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.1016. Min Act: -0.0001

Data Index: 822419 (C4 (Web Text))

Max Activating Token Index: 665

Text #8

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.1008. Min Act: -0.0001

Data Index: 1337398 (C4 (Web Text))

Max Activating Token Index: 674

Text #9

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.0990. Min Act: -0.0000

Data Index: 148500 (C4 (Web Text))

Max Activating Token Index: 817

Text #10

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.1024. Min Act: -0.0000

Data Index: 1261667 (C4 (Web Text))

Max Activating Token Index: 775

Text #11

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.1009. Min Act: -0.0000

Data Index: 159102 (C4 (Web Text))

Max Activating Token Index: 738

Text #12

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.0970. Min Act: -0.0000

Data Index: 588929 (C4 (Web Text))

Max Activating Token Index: 587

Text #13

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.0930. Min Act: -0.0000

Data Index: 1496365 (Python Code)

Max Activating Token Index: 306

Text #14

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.1041. Min Act: -0.0000

Data Index: 529776 (C4 (Web Text))

Max Activating Token Index: 583

Text #15

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.0927. Min Act: -0.0001

Data Index: 26148 (C4 (Web Text))

Max Activating Token Index: 787

Text #16

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.0953. Min Act: -0.0001

Data Index: 474218 (C4 (Web Text))

Max Activating Token Index: 62

Text #17

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.0983. Min Act: -0.0001

Data Index: 1286876 (C4 (Web Text))

Max Activating Token Index: 901

Text #18

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.0944. Min Act: -0.0001

Data Index: 475968 (C4 (Web Text))

Max Activating Token Index: 428

Text #19

Max Range: 0.1266. Min Range: -0.1266

Max Act: 0.0909. Min Act: -0.0000

Data Index: 479279 (C4 (Web Text))

Max Activating Token Index: 844