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

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

Neuron 1072 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.6339. Min Act: -0.0001

Data Index: 1474281 (Python Code)

Max Activating Token Index: 923

Text #1

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.6825. Min Act: -0.0001

Data Index: 523948 (C4 (Web Text))

Max Activating Token Index: 423

Text #2

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.5956. Min Act: -0.0001

Data Index: 1407831 (Python Code)

Max Activating Token Index: 35

Text #3

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.5077. Min Act: -0.0001

Data Index: 625313 (C4 (Web Text))

Max Activating Token Index: 928

Text #4

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.3924. Min Act: -0.0001

Data Index: 1042959 (C4 (Web Text))

Max Activating Token Index: 55

Text #5

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.3930. Min Act: -0.0001

Data Index: 1320780 (C4 (Web Text))

Max Activating Token Index: 662

Text #6

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.3535. Min Act: -0.0001

Data Index: 18455 (C4 (Web Text))

Max Activating Token Index: 594

Text #7

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.2846. Min Act: -0.0001

Data Index: 1344065 (C4 (Web Text))

Max Activating Token Index: 274

Text #8

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.3044. Min Act: -0.0001

Data Index: 657969 (C4 (Web Text))

Max Activating Token Index: 367

Text #9

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.3465. Min Act: -0.0001

Data Index: 1184863 (C4 (Web Text))

Max Activating Token Index: 348

Text #10

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.2175. Min Act: -0.0001

Data Index: 114419 (C4 (Web Text))

Max Activating Token Index: 683

Text #11

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.2610. Min Act: -0.0000

Data Index: 1391906 (Python Code)

Max Activating Token Index: 4

Text #12

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.2537. Min Act: -0.0001

Data Index: 1418200 (Python Code)

Max Activating Token Index: 978

Text #13

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.3006. Min Act: -0.0001

Data Index: 855115 (C4 (Web Text))

Max Activating Token Index: 604

Text #14

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.2267. Min Act: -0.0001

Data Index: 1212736 (C4 (Web Text))

Max Activating Token Index: 763

Text #15

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.0995. Min Act: -0.0001

Data Index: 1284540 (C4 (Web Text))

Max Activating Token Index: 809

Text #16

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.2902. Min Act: -0.0001

Data Index: 1213533 (C4 (Web Text))

Max Activating Token Index: 605

Text #17

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.2037. Min Act: -0.0001

Data Index: 1066325 (C4 (Web Text))

Max Activating Token Index: 681

Text #18

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.1207. Min Act: -0.0001

Data Index: 1475961 (Python Code)

Max Activating Token Index: 51

Text #19

Max Range: 1.6825. Min Range: -1.6825

Max Act: 1.2405. Min Act: -0.0001

Data Index: 200376 (C4 (Web Text))

Max Activating Token Index: 514