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

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

Neuron 2672 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.9813. Min Act: -0.0000

Data Index: 436504 (C4 (Web Text))

Max Activating Token Index: 246

Text #1

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.8568. Min Act: -0.0000

Data Index: 292099 (C4 (Web Text))

Max Activating Token Index: 162

Text #2

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.8268. Min Act: -0.0000

Data Index: 954419 (C4 (Web Text))

Max Activating Token Index: 549

Text #3

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.7830. Min Act: -0.0000

Data Index: 248 (C4 (Web Text))

Max Activating Token Index: 497

Text #4

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.7566. Min Act: -0.0000

Data Index: 4372 (C4 (Web Text))

Max Activating Token Index: 158

Text #5

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.7530. Min Act: -0.0000

Data Index: 1160227 (C4 (Web Text))

Max Activating Token Index: 544

Text #6

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.7214. Min Act: -0.0000

Data Index: 1250031 (C4 (Web Text))

Max Activating Token Index: 286

Text #7

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.6945. Min Act: -0.0000

Data Index: 181715 (C4 (Web Text))

Max Activating Token Index: 735

Text #8

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.6689. Min Act: -0.0000

Data Index: 212148 (C4 (Web Text))

Max Activating Token Index: 338

Text #9

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.6871. Min Act: -0.0000

Data Index: 1574871 (Python Code)

Max Activating Token Index: 853

Text #10

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.6556. Min Act: -0.0000

Data Index: 1651215 (Python Code)

Max Activating Token Index: 89

Text #11

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.6285. Min Act: -0.0000

Data Index: 1502764 (Python Code)

Max Activating Token Index: 520

Text #12

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.6263. Min Act: -0.0000

Data Index: 410098 (C4 (Web Text))

Max Activating Token Index: 149

Text #13

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.6144. Min Act: -0.0000

Data Index: 1455915 (Python Code)

Max Activating Token Index: 885

Text #14

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.6150. Min Act: -0.0000

Data Index: 1535697 (Python Code)

Max Activating Token Index: 115

Text #15

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.6063. Min Act: -0.0000

Data Index: 1649132 (Python Code)

Max Activating Token Index: 569

Text #16

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.5494. Min Act: -0.0000

Data Index: 403898 (C4 (Web Text))

Max Activating Token Index: 196

Text #17

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.5351. Min Act: -0.0000

Data Index: 240479 (C4 (Web Text))

Max Activating Token Index: 496

Text #18

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.5162. Min Act: -0.0000

Data Index: 1310551 (C4 (Web Text))

Max Activating Token Index: 465

Text #19

Max Range: 0.9813. Min Range: -0.9813

Max Act: 0.4990. Min Act: -0.0000

Data Index: 1495645 (Python Code)

Max Activating Token Index: 439