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

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

Neuron 2010 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.6186. Min Range: -2.6186

Max Act: 2.6186. Min Act: -0.0000

Data Index: 253403 (C4 (Web Text))

Max Activating Token Index: 2

Text #1

Max Range: 2.6186. Min Range: -2.6186

Max Act: 2.2332. Min Act: -0.0000

Data Index: 1448245 (Python Code)

Max Activating Token Index: 963

Text #2

Max Range: 2.6186. Min Range: -2.6186

Max Act: 2.1969. Min Act: -0.0000

Data Index: 1589774 (Python Code)

Max Activating Token Index: 913

Text #3

Max Range: 2.6186. Min Range: -2.6186

Max Act: 2.0450. Min Act: -0.0000

Data Index: 1607142 (Python Code)

Max Activating Token Index: 1015

Text #4

Max Range: 2.6186. Min Range: -2.6186

Max Act: 2.1088. Min Act: -0.0000

Data Index: 501544 (C4 (Web Text))

Max Activating Token Index: 718

Text #5

Max Range: 2.6186. Min Range: -2.6186

Max Act: 2.1096. Min Act: -0.0000

Data Index: 1365940 (Python Code)

Max Activating Token Index: 886

Text #6

Max Range: 2.6186. Min Range: -2.6186

Max Act: 2.0407. Min Act: -0.0000

Data Index: 1556380 (Python Code)

Max Activating Token Index: 150

Text #7

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.9104. Min Act: -0.0000

Data Index: 1426488 (Python Code)

Max Activating Token Index: 394

Text #8

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.7668. Min Act: -0.0000

Data Index: 1438876 (Python Code)

Max Activating Token Index: 49

Text #9

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.7709. Min Act: -0.0000

Data Index: 197399 (C4 (Web Text))

Max Activating Token Index: 259

Text #10

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.6701. Min Act: -0.0000

Data Index: 907646 (C4 (Web Text))

Max Activating Token Index: 1002

Text #11

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.7356. Min Act: -0.0000

Data Index: 1461739 (Python Code)

Max Activating Token Index: 60

Text #12

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.6098. Min Act: -0.0000

Data Index: 515300 (C4 (Web Text))

Max Activating Token Index: 929

Text #13

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.6858. Min Act: -0.0000

Data Index: 1481813 (Python Code)

Max Activating Token Index: 277

Text #14

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.6796. Min Act: -0.0000

Data Index: 1064425 (C4 (Web Text))

Max Activating Token Index: 263

Text #15

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.6303. Min Act: -0.0000

Data Index: 518138 (C4 (Web Text))

Max Activating Token Index: 121

Text #16

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.6255. Min Act: -0.0000

Data Index: 350473 (C4 (Web Text))

Max Activating Token Index: 573

Text #17

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.6010. Min Act: -0.0000

Data Index: 1153798 (C4 (Web Text))

Max Activating Token Index: 7

Text #18

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.5794. Min Act: -0.0000

Data Index: 1351451 (C4 (Web Text))

Max Activating Token Index: 911

Text #19

Max Range: 2.6186. Min Range: -2.6186

Max Act: 1.5807. Min Act: -0.0000

Data Index: 160545 (C4 (Web Text))

Max Activating Token Index: 70