Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 39 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.5473. Min Act: -0.0001

Data Index: 716051 (C4 (Web Text))

Max Activating Token Index: 397

Text #1

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.5066. Min Act: -0.0001

Data Index: 61723 (C4 (Web Text))

Max Activating Token Index: 589

Text #2

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4958. Min Act: -0.0001

Data Index: 907507 (C4 (Web Text))

Max Activating Token Index: 618

Text #3

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4557. Min Act: -0.0001

Data Index: 863767 (C4 (Web Text))

Max Activating Token Index: 355

Text #4

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4670. Min Act: -0.0001

Data Index: 838104 (C4 (Web Text))

Max Activating Token Index: 769

Text #5

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4537. Min Act: -0.0001

Data Index: 1056555 (C4 (Web Text))

Max Activating Token Index: 552

Text #6

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4469. Min Act: -0.0001

Data Index: 421648 (C4 (Web Text))

Max Activating Token Index: 868

Text #7

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4117. Min Act: -0.0001

Data Index: 1327008 (C4 (Web Text))

Max Activating Token Index: 571

Text #8

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4519. Min Act: 0.0000

Data Index: 792880 (C4 (Web Text))

Max Activating Token Index: 1000

Text #9

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4268. Min Act: -0.0001

Data Index: 856276 (C4 (Web Text))

Max Activating Token Index: 132

Text #10

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4363. Min Act: -0.0001

Data Index: 589159 (C4 (Web Text))

Max Activating Token Index: 750

Text #11

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4203. Min Act: -0.0001

Data Index: 39476 (C4 (Web Text))

Max Activating Token Index: 276

Text #12

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4320. Min Act: -0.0001

Data Index: 343570 (C4 (Web Text))

Max Activating Token Index: 310

Text #13

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4372. Min Act: -0.0001

Data Index: 428570 (C4 (Web Text))

Max Activating Token Index: 925

Text #14

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.4349. Min Act: -0.0000

Data Index: 721718 (C4 (Web Text))

Max Activating Token Index: 272

Text #15

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.3998. Min Act: -0.0001

Data Index: 148346 (C4 (Web Text))

Max Activating Token Index: 464

Text #16

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.3808. Min Act: -0.0001

Data Index: 606243 (C4 (Web Text))

Max Activating Token Index: 499

Text #17

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.3716. Min Act: -0.0000

Data Index: 745795 (C4 (Web Text))

Max Activating Token Index: 808

Text #18

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.3660. Min Act: -0.0001

Data Index: 283539 (C4 (Web Text))

Max Activating Token Index: 862

Text #19

Max Range: 0.5473. Min Range: -0.5473

Max Act: 0.3733. Min Act: 0.0001

Data Index: 859816 (C4 (Web Text))

Max Activating Token Index: 798