Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 1518 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.5046. Min Act: -0.0001

Data Index: 1200715 (C4 (Web Text))

Max Activating Token Index: 583

Text #1

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.5087. Min Act: -0.0001

Data Index: 1170969 (C4 (Web Text))

Max Activating Token Index: 110

Text #2

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.5089. Min Act: -0.0001

Data Index: 1536095 (Python Code)

Max Activating Token Index: 17

Text #3

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4933. Min Act: -0.0001

Data Index: 1313366 (C4 (Web Text))

Max Activating Token Index: 825

Text #4

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.5010. Min Act: -0.0001

Data Index: 831788 (C4 (Web Text))

Max Activating Token Index: 512

Text #5

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.5008. Min Act: -0.0001

Data Index: 512570 (C4 (Web Text))

Max Activating Token Index: 634

Text #6

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4172. Min Act: -0.0001

Data Index: 710260 (C4 (Web Text))

Max Activating Token Index: 671

Text #7

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4879. Min Act: -0.0001

Data Index: 1300075 (C4 (Web Text))

Max Activating Token Index: 648

Text #8

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4418. Min Act: -0.0001

Data Index: 984545 (C4 (Web Text))

Max Activating Token Index: 116

Text #9

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4267. Min Act: -0.0001

Data Index: 267708 (C4 (Web Text))

Max Activating Token Index: 834

Text #10

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4597. Min Act: -0.0001

Data Index: 1207487 (C4 (Web Text))

Max Activating Token Index: 997

Text #11

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4198. Min Act: -0.0001

Data Index: 290547 (C4 (Web Text))

Max Activating Token Index: 350

Text #12

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4711. Min Act: -0.0001

Data Index: 888247 (C4 (Web Text))

Max Activating Token Index: 190

Text #13

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4461. Min Act: -0.0001

Data Index: 774684 (C4 (Web Text))

Max Activating Token Index: 287

Text #14

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4457. Min Act: -0.0001

Data Index: 527669 (C4 (Web Text))

Max Activating Token Index: 382

Text #15

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4337. Min Act: -0.0001

Data Index: 692488 (C4 (Web Text))

Max Activating Token Index: 107

Text #16

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4398. Min Act: -0.0001

Data Index: 179429 (C4 (Web Text))

Max Activating Token Index: 375

Text #17

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4266. Min Act: -0.0001

Data Index: 405700 (C4 (Web Text))

Max Activating Token Index: 407

Text #18

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4116. Min Act: -0.0001

Data Index: 579003 (C4 (Web Text))

Max Activating Token Index: 280

Text #19

Max Range: 0.5089. Min Range: -0.5089

Max Act: 0.4395. Min Act: -0.0001

Data Index: 11616 (C4 (Web Text))

Max Activating Token Index: 376