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

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

Neuron 1633 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.1981. Min Range: -1.1981

Max Act: 1.0107. Min Act: -0.0001

Data Index: 19347 (C4 (Web Text))

Max Activating Token Index: 218

Text #1

Max Range: 1.1981. Min Range: -1.1981

Max Act: 1.1981. Min Act: -0.0001

Data Index: 789606 (C4 (Web Text))

Max Activating Token Index: 690

Text #2

Max Range: 1.1981. Min Range: -1.1981

Max Act: 1.0071. Min Act: -0.0001

Data Index: 3376 (C4 (Web Text))

Max Activating Token Index: 10

Text #3

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.8519. Min Act: -0.0001

Data Index: 1112065 (C4 (Web Text))

Max Activating Token Index: 577

Text #4

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.8978. Min Act: -0.0001

Data Index: 874401 (C4 (Web Text))

Max Activating Token Index: 465

Text #5

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.8858. Min Act: -0.0001

Data Index: 135830 (C4 (Web Text))

Max Activating Token Index: 273

Text #6

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.8009. Min Act: -0.0001

Data Index: 237389 (C4 (Web Text))

Max Activating Token Index: 847

Text #7

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.8906. Min Act: -0.0001

Data Index: 1516468 (Python Code)

Max Activating Token Index: 363

Text #8

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.8353. Min Act: -0.0001

Data Index: 497079 (C4 (Web Text))

Max Activating Token Index: 484

Text #9

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.7850. Min Act: -0.0001

Data Index: 1346034 (C4 (Web Text))

Max Activating Token Index: 524

Text #10

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.7372. Min Act: -0.0001

Data Index: 19657 (C4 (Web Text))

Max Activating Token Index: 507

Text #11

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.8023. Min Act: -0.0001

Data Index: 676619 (C4 (Web Text))

Max Activating Token Index: 543

Text #12

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.8325. Min Act: -0.0001

Data Index: 812432 (C4 (Web Text))

Max Activating Token Index: 48

Text #13

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.7968. Min Act: -0.0001

Data Index: 690402 (C4 (Web Text))

Max Activating Token Index: 759

Text #14

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.7992. Min Act: -0.0001

Data Index: 1438685 (Python Code)

Max Activating Token Index: 115

Text #15

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.7913. Min Act: -0.0001

Data Index: 542989 (C4 (Web Text))

Max Activating Token Index: 23

Text #16

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.8353. Min Act: -0.0001

Data Index: 1636420 (Python Code)

Max Activating Token Index: 613

Text #17

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.7771. Min Act: -0.0001

Data Index: 719581 (C4 (Web Text))

Max Activating Token Index: 211

Text #18

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.7517. Min Act: -0.0001

Data Index: 514954 (C4 (Web Text))

Max Activating Token Index: 27

Text #19

Max Range: 1.1981. Min Range: -1.1981

Max Act: 0.7978. Min Act: -0.0001

Data Index: 371082 (C4 (Web Text))

Max Activating Token Index: 27