Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3602 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.0108. Min Range: -1.0108

Max Act: 1.0108. Min Act: -0.0000

Data Index: 1620397 (Python Code)

Max Activating Token Index: 100

Text #1

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.9452. Min Act: -0.0000

Data Index: 1176504 (C4 (Web Text))

Max Activating Token Index: 133

Text #2

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.9968. Min Act: -0.0000

Data Index: 1550399 (Python Code)

Max Activating Token Index: 677

Text #3

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.9188. Min Act: -0.0000

Data Index: 1557633 (Python Code)

Max Activating Token Index: 44

Text #4

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.8657. Min Act: -0.0000

Data Index: 512006 (C4 (Web Text))

Max Activating Token Index: 433

Text #5

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.8487. Min Act: -0.0000

Data Index: 757827 (C4 (Web Text))

Max Activating Token Index: 168

Text #6

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.8664. Min Act: -0.0000

Data Index: 1563746 (Python Code)

Max Activating Token Index: 101

Text #7

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.8326. Min Act: -0.0000

Data Index: 569324 (C4 (Web Text))

Max Activating Token Index: 156

Text #8

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.8078. Min Act: -0.0000

Data Index: 1229901 (C4 (Web Text))

Max Activating Token Index: 469

Text #9

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.8030. Min Act: -0.0000

Data Index: 1184455 (C4 (Web Text))

Max Activating Token Index: 613

Text #10

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.8411. Min Act: -0.0000

Data Index: 208279 (C4 (Web Text))

Max Activating Token Index: 315

Text #11

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.8262. Min Act: -0.0000

Data Index: 1437811 (Python Code)

Max Activating Token Index: 679

Text #12

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.7622. Min Act: -0.0000

Data Index: 501665 (C4 (Web Text))

Max Activating Token Index: 643

Text #13

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.7596. Min Act: -0.0000

Data Index: 829041 (C4 (Web Text))

Max Activating Token Index: 858

Text #14

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.7930. Min Act: -0.0000

Data Index: 84408 (C4 (Web Text))

Max Activating Token Index: 881

Text #15

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.7646. Min Act: -0.0000

Data Index: 952558 (C4 (Web Text))

Max Activating Token Index: 482

Text #16

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.7566. Min Act: -0.0000

Data Index: 225559 (C4 (Web Text))

Max Activating Token Index: 724

Text #17

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.7862. Min Act: -0.0000

Data Index: 1096191 (C4 (Web Text))

Max Activating Token Index: 441

Text #18

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.7120. Min Act: -0.0000

Data Index: 1365320 (Python Code)

Max Activating Token Index: 682

Text #19

Max Range: 1.0108. Min Range: -1.0108

Max Act: 0.7057. Min Act: -0.0000

Data Index: 342978 (C4 (Web Text))

Max Activating Token Index: 649