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

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

Neuron 4591 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: 0.0807. Min Range: -0.0807

Max Act: 0.0807. Min Act: -0.0000

Data Index: 1543362 (Python Code)

Max Activating Token Index: 502

Text #1

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0757. Min Act: -0.0000

Data Index: 67242 (C4 (Web Text))

Max Activating Token Index: 560

Text #2

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0730. Min Act: -0.0000

Data Index: 307822 (C4 (Web Text))

Max Activating Token Index: 391

Text #3

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0593. Min Act: -0.0000

Data Index: 343318 (C4 (Web Text))

Max Activating Token Index: 332

Text #4

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0580. Min Act: -0.0000

Data Index: 1257503 (C4 (Web Text))

Max Activating Token Index: 437

Text #5

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0644. Min Act: -0.0000

Data Index: 815548 (C4 (Web Text))

Max Activating Token Index: 773

Text #6

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0617. Min Act: -0.0000

Data Index: 634750 (C4 (Web Text))

Max Activating Token Index: 154

Text #7

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0579. Min Act: -0.0000

Data Index: 305932 (C4 (Web Text))

Max Activating Token Index: 534

Text #8

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0591. Min Act: -0.0000

Data Index: 365936 (C4 (Web Text))

Max Activating Token Index: 510

Text #9

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0554. Min Act: -0.0000

Data Index: 592511 (C4 (Web Text))

Max Activating Token Index: 610

Text #10

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0530. Min Act: -0.0000

Data Index: 1066025 (C4 (Web Text))

Max Activating Token Index: 882

Text #11

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0540. Min Act: -0.0000

Data Index: 364313 (C4 (Web Text))

Max Activating Token Index: 923

Text #12

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0533. Min Act: -0.0000

Data Index: 887703 (C4 (Web Text))

Max Activating Token Index: 999

Text #13

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0560. Min Act: -0.0000

Data Index: 520660 (C4 (Web Text))

Max Activating Token Index: 186

Text #14

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0521. Min Act: -0.0000

Data Index: 684048 (C4 (Web Text))

Max Activating Token Index: 935

Text #15

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0525. Min Act: -0.0000

Data Index: 1465013 (Python Code)

Max Activating Token Index: 241

Text #16

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0568. Min Act: -0.0000

Data Index: 1118001 (C4 (Web Text))

Max Activating Token Index: 298

Text #17

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0511. Min Act: -0.0000

Data Index: 1295992 (C4 (Web Text))

Max Activating Token Index: 887

Text #18

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0521. Min Act: -0.0000

Data Index: 988268 (C4 (Web Text))

Max Activating Token Index: 222

Text #19

Max Range: 0.0807. Min Range: -0.0807

Max Act: 0.0538. Min Act: -0.0000

Data Index: 847413 (C4 (Web Text))

Max Activating Token Index: 724