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

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

Neuron 3480 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: 0.0861. Min Range: -0.0861

Max Act: 0.0861. Min Act: -0.0001

Data Index: 1324190 (C4 (Web Text))

Max Activating Token Index: 673

Text #1

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0826. Min Act: -0.0001

Data Index: 147798 (C4 (Web Text))

Max Activating Token Index: 696

Text #2

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0723. Min Act: -0.0001

Data Index: 1245281 (C4 (Web Text))

Max Activating Token Index: 467

Text #3

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0565. Min Act: -0.0000

Data Index: 33881 (C4 (Web Text))

Max Activating Token Index: 433

Text #4

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0505. Min Act: -0.0001

Data Index: 921227 (C4 (Web Text))

Max Activating Token Index: 769

Text #5

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0528. Min Act: -0.0000

Data Index: 770031 (C4 (Web Text))

Max Activating Token Index: 242

Text #6

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0477. Min Act: -0.0001

Data Index: 387068 (C4 (Web Text))

Max Activating Token Index: 775

Text #7

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0409. Min Act: -0.0001

Data Index: 14902 (C4 (Web Text))

Max Activating Token Index: 347

Text #8

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0460. Min Act: -0.0001

Data Index: 1444989 (Python Code)

Max Activating Token Index: 1012

Text #9

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0449. Min Act: -0.0001

Data Index: 125876 (C4 (Web Text))

Max Activating Token Index: 570

Text #10

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0445. Min Act: -0.0001

Data Index: 514553 (C4 (Web Text))

Max Activating Token Index: 349

Text #11

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0420. Min Act: -0.0001

Data Index: 932664 (C4 (Web Text))

Max Activating Token Index: 812

Text #12

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0393. Min Act: -0.0000

Data Index: 889623 (C4 (Web Text))

Max Activating Token Index: 632

Text #13

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0401. Min Act: -0.0001

Data Index: 79616 (C4 (Web Text))

Max Activating Token Index: 891

Text #14

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0405. Min Act: -0.0001

Data Index: 1611496 (Python Code)

Max Activating Token Index: 909

Text #15

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0380. Min Act: -0.0000

Data Index: 1295420 (C4 (Web Text))

Max Activating Token Index: 87

Text #16

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0385. Min Act: -0.0001

Data Index: 324049 (C4 (Web Text))

Max Activating Token Index: 608

Text #17

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0358. Min Act: -0.0001

Data Index: 357798 (C4 (Web Text))

Max Activating Token Index: 536

Text #18

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0370. Min Act: -0.0001

Data Index: 900202 (C4 (Web Text))

Max Activating Token Index: 660

Text #19

Max Range: 0.0861. Min Range: -0.0861

Max Act: 0.0374. Min Act: -0.0000

Data Index: 558735 (C4 (Web Text))

Max Activating Token Index: 82