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

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

Neuron 3820 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.4937. Min Act: -0.0001

Data Index: 1085924 (C4 (Web Text))

Max Activating Token Index: 895

Text #1

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.4656. Min Act: -0.0001

Data Index: 1375784 (Python Code)

Max Activating Token Index: 1022

Text #2

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3888. Min Act: -0.0001

Data Index: 1556780 (Python Code)

Max Activating Token Index: 1000

Text #3

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.4170. Min Act: -0.0001

Data Index: 80977 (C4 (Web Text))

Max Activating Token Index: 681

Text #4

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3929. Min Act: -0.0001

Data Index: 839494 (C4 (Web Text))

Max Activating Token Index: 872

Text #5

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3807. Min Act: -0.0001

Data Index: 307392 (C4 (Web Text))

Max Activating Token Index: 542

Text #6

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3837. Min Act: -0.0001

Data Index: 1468175 (Python Code)

Max Activating Token Index: 1015

Text #7

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3815. Min Act: -0.0001

Data Index: 348919 (C4 (Web Text))

Max Activating Token Index: 925

Text #8

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3605. Min Act: -0.0001

Data Index: 281950 (C4 (Web Text))

Max Activating Token Index: 579

Text #9

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3779. Min Act: -0.0001

Data Index: 1634043 (Python Code)

Max Activating Token Index: 786

Text #10

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3282. Min Act: -0.0001

Data Index: 1330655 (C4 (Web Text))

Max Activating Token Index: 761

Text #11

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3585. Min Act: -0.0001

Data Index: 980539 (C4 (Web Text))

Max Activating Token Index: 921

Text #12

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3552. Min Act: -0.0001

Data Index: 1212257 (C4 (Web Text))

Max Activating Token Index: 459

Text #13

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3324. Min Act: -0.0001

Data Index: 1589349 (Python Code)

Max Activating Token Index: 620

Text #14

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3442. Min Act: -0.0001

Data Index: 614072 (C4 (Web Text))

Max Activating Token Index: 896

Text #15

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3650. Min Act: -0.0001

Data Index: 1393008 (Python Code)

Max Activating Token Index: 913

Text #16

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3551. Min Act: -0.0001

Data Index: 1437568 (Python Code)

Max Activating Token Index: 650

Text #17

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3599. Min Act: -0.0001

Data Index: 1382521 (Python Code)

Max Activating Token Index: 932

Text #18

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3353. Min Act: -0.0001

Data Index: 1399604 (Python Code)

Max Activating Token Index: 847

Text #19

Max Range: 0.4937. Min Range: -0.4937

Max Act: 0.3283. Min Act: -0.0001

Data Index: 1473835 (Python Code)

Max Activating Token Index: 884