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

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

Neuron 187 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: 0.4825. Min Range: -0.4825

Max Act: 0.4825. Min Act: -0.0001

Data Index: 802428 (C4 (Web Text))

Max Activating Token Index: 588

Text #1

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.4778. Min Act: -0.0001

Data Index: 1327977 (C4 (Web Text))

Max Activating Token Index: 168

Text #2

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.4457. Min Act: -0.0001

Data Index: 9327 (C4 (Web Text))

Max Activating Token Index: 142

Text #3

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3869. Min Act: -0.0001

Data Index: 419063 (C4 (Web Text))

Max Activating Token Index: 749

Text #4

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3734. Min Act: -0.0001

Data Index: 446213 (C4 (Web Text))

Max Activating Token Index: 372

Text #5

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3610. Min Act: -0.0001

Data Index: 84517 (C4 (Web Text))

Max Activating Token Index: 832

Text #6

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3606. Min Act: -0.0001

Data Index: 1085249 (C4 (Web Text))

Max Activating Token Index: 996

Text #7

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3546. Min Act: -0.0001

Data Index: 1202745 (C4 (Web Text))

Max Activating Token Index: 768

Text #8

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3399. Min Act: -0.0001

Data Index: 467300 (C4 (Web Text))

Max Activating Token Index: 770

Text #9

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3236. Min Act: -0.0001

Data Index: 1609933 (Python Code)

Max Activating Token Index: 436

Text #10

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3420. Min Act: -0.0001

Data Index: 501787 (C4 (Web Text))

Max Activating Token Index: 281

Text #11

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3308. Min Act: -0.0001

Data Index: 776270 (C4 (Web Text))

Max Activating Token Index: 1009

Text #12

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3149. Min Act: -0.0001

Data Index: 426285 (C4 (Web Text))

Max Activating Token Index: 53

Text #13

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3217. Min Act: -0.0001

Data Index: 578048 (C4 (Web Text))

Max Activating Token Index: 501

Text #14

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3313. Min Act: -0.0001

Data Index: 744975 (C4 (Web Text))

Max Activating Token Index: 437

Text #15

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3225. Min Act: -0.0001

Data Index: 1097345 (C4 (Web Text))

Max Activating Token Index: 414

Text #16

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3173. Min Act: -0.0001

Data Index: 1331564 (C4 (Web Text))

Max Activating Token Index: 80

Text #17

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3176. Min Act: -0.0001

Data Index: 1347619 (C4 (Web Text))

Max Activating Token Index: 59

Text #18

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3355. Min Act: -0.0001

Data Index: 706952 (C4 (Web Text))

Max Activating Token Index: 238

Text #19

Max Range: 0.4825. Min Range: -0.4825

Max Act: 0.3047. Min Act: -0.0001

Data Index: 1377807 (Python Code)

Max Activating Token Index: 516