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

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

Neuron 1138 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.7715. Min Act: -0.0000

Data Index: 486479 (C4 (Web Text))

Max Activating Token Index: 702

Text #1

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.6252. Min Act: -0.0000

Data Index: 104580 (C4 (Web Text))

Max Activating Token Index: 774

Text #2

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.5105. Min Act: -0.0000

Data Index: 950668 (C4 (Web Text))

Max Activating Token Index: 235

Text #3

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.5382. Min Act: -0.0000

Data Index: 759517 (C4 (Web Text))

Max Activating Token Index: 63

Text #4

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.5149. Min Act: -0.0000

Data Index: 659364 (C4 (Web Text))

Max Activating Token Index: 391

Text #5

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.5045. Min Act: -0.0000

Data Index: 66590 (C4 (Web Text))

Max Activating Token Index: 62

Text #6

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4982. Min Act: -0.0000

Data Index: 1106276 (C4 (Web Text))

Max Activating Token Index: 834

Text #7

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4426. Min Act: -0.0000

Data Index: 312721 (C4 (Web Text))

Max Activating Token Index: 209

Text #8

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4894. Min Act: -0.0000

Data Index: 1517290 (Python Code)

Max Activating Token Index: 724

Text #9

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4564. Min Act: -0.0000

Data Index: 1186619 (C4 (Web Text))

Max Activating Token Index: 230

Text #10

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4443. Min Act: -0.0000

Data Index: 1313620 (C4 (Web Text))

Max Activating Token Index: 849

Text #11

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4704. Min Act: -0.0000

Data Index: 1232973 (C4 (Web Text))

Max Activating Token Index: 64

Text #12

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4562. Min Act: -0.0000

Data Index: 1458402 (Python Code)

Max Activating Token Index: 107

Text #13

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4498. Min Act: -0.0000

Data Index: 1062924 (C4 (Web Text))

Max Activating Token Index: 104

Text #14

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4421. Min Act: -0.0000

Data Index: 1054577 (C4 (Web Text))

Max Activating Token Index: 478

Text #15

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4965. Min Act: -0.0000

Data Index: 744607 (C4 (Web Text))

Max Activating Token Index: 700

Text #16

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4407. Min Act: -0.0000

Data Index: 1383044 (Python Code)

Max Activating Token Index: 126

Text #17

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4548. Min Act: -0.0000

Data Index: 992828 (C4 (Web Text))

Max Activating Token Index: 383

Text #18

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4547. Min Act: -0.0000

Data Index: 709853 (C4 (Web Text))

Max Activating Token Index: 158

Text #19

Max Range: 0.7715. Min Range: -0.7715

Max Act: 0.4743. Min Act: -0.0000

Data Index: 1160929 (C4 (Web Text))

Max Activating Token Index: 717