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

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

Neuron 2630 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.8540. Min Act: -0.0000

Data Index: 966796 (C4 (Web Text))

Max Activating Token Index: 667

Text #1

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.8524. Min Act: -0.0000

Data Index: 545573 (C4 (Web Text))

Max Activating Token Index: 896

Text #2

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.8081. Min Act: -0.0000

Data Index: 757222 (C4 (Web Text))

Max Activating Token Index: 156

Text #3

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.6980. Min Act: -0.0000

Data Index: 1002039 (C4 (Web Text))

Max Activating Token Index: 476

Text #4

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.6741. Min Act: -0.0000

Data Index: 125456 (C4 (Web Text))

Max Activating Token Index: 966

Text #5

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.5962. Min Act: -0.0000

Data Index: 323583 (C4 (Web Text))

Max Activating Token Index: 340

Text #6

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.5279. Min Act: -0.0000

Data Index: 1489779 (Python Code)

Max Activating Token Index: 701

Text #7

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.5977. Min Act: -0.0000

Data Index: 826527 (C4 (Web Text))

Max Activating Token Index: 41

Text #8

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.5125. Min Act: -0.0000

Data Index: 616818 (C4 (Web Text))

Max Activating Token Index: 248

Text #9

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.5640. Min Act: -0.0000

Data Index: 61257 (C4 (Web Text))

Max Activating Token Index: 1004

Text #10

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.5663. Min Act: -0.0000

Data Index: 534002 (C4 (Web Text))

Max Activating Token Index: 137

Text #11

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.4292. Min Act: -0.0000

Data Index: 403362 (C4 (Web Text))

Max Activating Token Index: 473

Text #12

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.4339. Min Act: -0.0000

Data Index: 163773 (C4 (Web Text))

Max Activating Token Index: 510

Text #13

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.4909. Min Act: -0.0000

Data Index: 1028623 (C4 (Web Text))

Max Activating Token Index: 77

Text #14

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.4049. Min Act: -0.0000

Data Index: 1523985 (Python Code)

Max Activating Token Index: 300

Text #15

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.4132. Min Act: -0.0000

Data Index: 85400 (C4 (Web Text))

Max Activating Token Index: 876

Text #16

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.3585. Min Act: -0.0000

Data Index: 1452379 (Python Code)

Max Activating Token Index: 218

Text #17

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.3910. Min Act: -0.0000

Data Index: 478826 (C4 (Web Text))

Max Activating Token Index: 160

Text #18

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.3278. Min Act: -0.0000

Data Index: 86991 (C4 (Web Text))

Max Activating Token Index: 578

Text #19

Max Range: 1.8540. Min Range: -1.8540

Max Act: 1.3188. Min Act: -0.0000

Data Index: 1329749 (C4 (Web Text))

Max Activating Token Index: 556