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

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

Neuron 2520 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.6541. Min Act: -0.0000

Data Index: 155934 (C4 (Web Text))

Max Activating Token Index: 172

Text #1

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.5773. Min Act: -0.0000

Data Index: 381594 (C4 (Web Text))

Max Activating Token Index: 576

Text #2

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.5302. Min Act: -0.0001

Data Index: 886044 (C4 (Web Text))

Max Activating Token Index: 148

Text #3

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.5793. Min Act: -0.0000

Data Index: 216714 (C4 (Web Text))

Max Activating Token Index: 751

Text #4

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.5440. Min Act: -0.0001

Data Index: 697480 (C4 (Web Text))

Max Activating Token Index: 245

Text #5

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4843. Min Act: -0.0000

Data Index: 924385 (C4 (Web Text))

Max Activating Token Index: 488

Text #6

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4612. Min Act: -0.0001

Data Index: 754159 (C4 (Web Text))

Max Activating Token Index: 573

Text #7

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4488. Min Act: -0.0001

Data Index: 1350443 (C4 (Web Text))

Max Activating Token Index: 331

Text #8

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4906. Min Act: -0.0001

Data Index: 1175249 (C4 (Web Text))

Max Activating Token Index: 615

Text #9

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4342. Min Act: -0.0000

Data Index: 1133451 (C4 (Web Text))

Max Activating Token Index: 710

Text #10

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4352. Min Act: -0.0001

Data Index: 685991 (C4 (Web Text))

Max Activating Token Index: 724

Text #11

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4790. Min Act: -0.0000

Data Index: 1393100 (Python Code)

Max Activating Token Index: 613

Text #12

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4068. Min Act: -0.0001

Data Index: 101493 (C4 (Web Text))

Max Activating Token Index: 381

Text #13

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4549. Min Act: -0.0000

Data Index: 986501 (C4 (Web Text))

Max Activating Token Index: 716

Text #14

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4337. Min Act: -0.0000

Data Index: 1126779 (C4 (Web Text))

Max Activating Token Index: 711

Text #15

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4198. Min Act: -0.0000

Data Index: 955826 (C4 (Web Text))

Max Activating Token Index: 839

Text #16

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4961. Min Act: -0.0001

Data Index: 604823 (C4 (Web Text))

Max Activating Token Index: 313

Text #17

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4361. Min Act: -0.0000

Data Index: 907114 (C4 (Web Text))

Max Activating Token Index: 747

Text #18

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4716. Min Act: -0.0000

Data Index: 222414 (C4 (Web Text))

Max Activating Token Index: 816

Text #19

Max Range: 0.6541. Min Range: -0.6541

Max Act: 0.4667. Min Act: -0.0001

Data Index: 1125387 (C4 (Web Text))

Max Activating Token Index: 666