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

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

Neuron 5239 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.3089. Min Act: -0.0001

Data Index: 768265 (C4 (Web Text))

Max Activating Token Index: 46

Text #1

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2861. Min Act: -0.0001

Data Index: 1017873 (C4 (Web Text))

Max Activating Token Index: 9

Text #2

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2555. Min Act: -0.0001

Data Index: 826932 (C4 (Web Text))

Max Activating Token Index: 42

Text #3

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2576. Min Act: -0.0001

Data Index: 982962 (C4 (Web Text))

Max Activating Token Index: 71

Text #4

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2320. Min Act: -0.0001

Data Index: 1355529 (C4 (Web Text))

Max Activating Token Index: 94

Text #5

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2448. Min Act: -0.0001

Data Index: 829264 (C4 (Web Text))

Max Activating Token Index: 671

Text #6

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2307. Min Act: -0.0001

Data Index: 326342 (C4 (Web Text))

Max Activating Token Index: 826

Text #7

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2254. Min Act: -0.0001

Data Index: 669006 (C4 (Web Text))

Max Activating Token Index: 275

Text #8

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.1783. Min Act: -0.0001

Data Index: 192795 (C4 (Web Text))

Max Activating Token Index: 275

Text #9

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2371. Min Act: -0.0001

Data Index: 397840 (C4 (Web Text))

Max Activating Token Index: 43

Text #10

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2177. Min Act: -0.0001

Data Index: 423560 (C4 (Web Text))

Max Activating Token Index: 849

Text #11

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2199. Min Act: -0.0001

Data Index: 307912 (C4 (Web Text))

Max Activating Token Index: 559

Text #12

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2275. Min Act: -0.0001

Data Index: 1290304 (C4 (Web Text))

Max Activating Token Index: 387

Text #13

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2304. Min Act: -0.0001

Data Index: 1026135 (C4 (Web Text))

Max Activating Token Index: 91

Text #14

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2082. Min Act: -0.0001

Data Index: 789145 (C4 (Web Text))

Max Activating Token Index: 46

Text #15

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.1920. Min Act: -0.0001

Data Index: 710903 (C4 (Web Text))

Max Activating Token Index: 1014

Text #16

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2105. Min Act: -0.0001

Data Index: 586988 (C4 (Web Text))

Max Activating Token Index: 422

Text #17

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.1916. Min Act: -0.0001

Data Index: 817118 (C4 (Web Text))

Max Activating Token Index: 527

Text #18

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.1987. Min Act: -0.0001

Data Index: 1450702 (Python Code)

Max Activating Token Index: 209

Text #19

Max Range: 0.3089. Min Range: -0.3089

Max Act: 0.2024. Min Act: -0.0001

Data Index: 468493 (C4 (Web Text))

Max Activating Token Index: 629