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

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

Neuron 1000 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: 0.8715. Min Range: -0.8715

Max Act: 0.8715. Min Act: -0.0001

Data Index: 816042 (C4 (Web Text))

Max Activating Token Index: 523

Text #1

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.8168. Min Act: -0.0001

Data Index: 521430 (C4 (Web Text))

Max Activating Token Index: 126

Text #2

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.8039. Min Act: -0.0000

Data Index: 110667 (C4 (Web Text))

Max Activating Token Index: 568

Text #3

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.7731. Min Act: -0.0000

Data Index: 553668 (C4 (Web Text))

Max Activating Token Index: 108

Text #4

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.6692. Min Act: -0.0000

Data Index: 632466 (C4 (Web Text))

Max Activating Token Index: 467

Text #5

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.6506. Min Act: -0.0001

Data Index: 803931 (C4 (Web Text))

Max Activating Token Index: 387

Text #6

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.6250. Min Act: -0.0000

Data Index: 829701 (C4 (Web Text))

Max Activating Token Index: 602

Text #7

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.6295. Min Act: -0.0000

Data Index: 671436 (C4 (Web Text))

Max Activating Token Index: 385

Text #8

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.6055. Min Act: -0.0000

Data Index: 125358 (C4 (Web Text))

Max Activating Token Index: 930

Text #9

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.6325. Min Act: -0.0000

Data Index: 650581 (C4 (Web Text))

Max Activating Token Index: 40

Text #10

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.5868. Min Act: -0.0000

Data Index: 669353 (C4 (Web Text))

Max Activating Token Index: 733

Text #11

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.5994. Min Act: -0.0000

Data Index: 922261 (C4 (Web Text))

Max Activating Token Index: 229

Text #12

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.5765. Min Act: -0.0000

Data Index: 588031 (C4 (Web Text))

Max Activating Token Index: 533

Text #13

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.6004. Min Act: -0.0000

Data Index: 827104 (C4 (Web Text))

Max Activating Token Index: 758

Text #14

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.5702. Min Act: -0.0000

Data Index: 448001 (C4 (Web Text))

Max Activating Token Index: 816

Text #15

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.5424. Min Act: -0.0000

Data Index: 1379841 (Python Code)

Max Activating Token Index: 712

Text #16

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.5650. Min Act: -0.0001

Data Index: 1558792 (Python Code)

Max Activating Token Index: 911

Text #17

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.5396. Min Act: -0.0000

Data Index: 424836 (C4 (Web Text))

Max Activating Token Index: 818

Text #18

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.5542. Min Act: -0.0001

Data Index: 690954 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 0.8715. Min Range: -0.8715

Max Act: 0.5552. Min Act: -0.0000

Data Index: 107228 (C4 (Web Text))

Max Activating Token Index: 545