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

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

Neuron 4795 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: 0.8554. Min Range: -0.8554

Max Act: 0.8323. Min Act: -0.0000

Data Index: 1087734 (C4 (Web Text))

Max Activating Token Index: 965

Text #1

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.7733. Min Act: -0.0000

Data Index: 156350 (C4 (Web Text))

Max Activating Token Index: 800

Text #2

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.8138. Min Act: -0.0000

Data Index: 547007 (C4 (Web Text))

Max Activating Token Index: 204

Text #3

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.7976. Min Act: -0.0000

Data Index: 1375443 (Python Code)

Max Activating Token Index: 815

Text #4

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.7813. Min Act: -0.0000

Data Index: 1460219 (Python Code)

Max Activating Token Index: 977

Text #5

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.8278. Min Act: -0.0000

Data Index: 911187 (C4 (Web Text))

Max Activating Token Index: 426

Text #6

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.7682. Min Act: -0.0000

Data Index: 197960 (C4 (Web Text))

Max Activating Token Index: 729

Text #7

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.7239. Min Act: -0.0000

Data Index: 1121810 (C4 (Web Text))

Max Activating Token Index: 258

Text #8

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.7638. Min Act: -0.0000

Data Index: 1422981 (Python Code)

Max Activating Token Index: 643

Text #9

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.7593. Min Act: -0.0000

Data Index: 228202 (C4 (Web Text))

Max Activating Token Index: 80

Text #10

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.7754. Min Act: -0.0000

Data Index: 1404527 (Python Code)

Max Activating Token Index: 324

Text #11

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.7319. Min Act: -0.0000

Data Index: 1412167 (Python Code)

Max Activating Token Index: 648

Text #12

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.7438. Min Act: -0.0000

Data Index: 1554646 (Python Code)

Max Activating Token Index: 395

Text #13

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.6353. Min Act: -0.0000

Data Index: 1560443 (Python Code)

Max Activating Token Index: 361

Text #14

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.6696. Min Act: -0.0000

Data Index: 1077067 (C4 (Web Text))

Max Activating Token Index: 848

Text #15

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.6878. Min Act: -0.0000

Data Index: 1517002 (Python Code)

Max Activating Token Index: 954

Text #16

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.7007. Min Act: -0.0000

Data Index: 528727 (C4 (Web Text))

Max Activating Token Index: 30

Text #17

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.7207. Min Act: -0.0000

Data Index: 325663 (C4 (Web Text))

Max Activating Token Index: 296

Text #18

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.8554. Min Act: -0.0000

Data Index: 821256 (C4 (Web Text))

Max Activating Token Index: 715

Text #19

Max Range: 0.8554. Min Range: -0.8554

Max Act: 0.6941. Min Act: -0.0000

Data Index: 580027 (C4 (Web Text))

Max Activating Token Index: 374