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

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

Neuron 2450 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: 0.8055. Min Range: -0.8055

Max Act: 0.8055. Min Act: -0.0000

Data Index: 751621 (C4 (Web Text))

Max Activating Token Index: 702

Text #1

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.6101. Min Act: -0.0000

Data Index: 1592483 (Python Code)

Max Activating Token Index: 616

Text #2

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.4961. Min Act: -0.0000

Data Index: 342737 (C4 (Web Text))

Max Activating Token Index: 603

Text #3

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.5353. Min Act: -0.0000

Data Index: 186285 (C4 (Web Text))

Max Activating Token Index: 832

Text #4

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.4621. Min Act: -0.0000

Data Index: 358712 (C4 (Web Text))

Max Activating Token Index: 1001

Text #5

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.4768. Min Act: -0.0000

Data Index: 490516 (C4 (Web Text))

Max Activating Token Index: 883

Text #6

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.4933. Min Act: -0.0000

Data Index: 27260 (C4 (Web Text))

Max Activating Token Index: 876

Text #7

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.4253. Min Act: -0.0000

Data Index: 1227568 (C4 (Web Text))

Max Activating Token Index: 318

Text #8

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.4679. Min Act: -0.0000

Data Index: 353005 (C4 (Web Text))

Max Activating Token Index: 889

Text #9

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.4798. Min Act: -0.0000

Data Index: 1135561 (C4 (Web Text))

Max Activating Token Index: 811

Text #10

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.4337. Min Act: -0.0000

Data Index: 550587 (C4 (Web Text))

Max Activating Token Index: 244

Text #11

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.4535. Min Act: -0.0001

Data Index: 1514929 (Python Code)

Max Activating Token Index: 643

Text #12

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.4174. Min Act: -0.0000

Data Index: 357427 (C4 (Web Text))

Max Activating Token Index: 974

Text #13

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.3965. Min Act: -0.0000

Data Index: 566736 (C4 (Web Text))

Max Activating Token Index: 115

Text #14

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.3885. Min Act: -0.0000

Data Index: 1595171 (Python Code)

Max Activating Token Index: 450

Text #15

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.3823. Min Act: -0.0000

Data Index: 1559297 (Python Code)

Max Activating Token Index: 983

Text #16

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.4005. Min Act: -0.0000

Data Index: 1589883 (Python Code)

Max Activating Token Index: 232

Text #17

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.3882. Min Act: -0.0000

Data Index: 42421 (C4 (Web Text))

Max Activating Token Index: 810

Text #18

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.3715. Min Act: -0.0000

Data Index: 339531 (C4 (Web Text))

Max Activating Token Index: 532

Text #19

Max Range: 0.8055. Min Range: -0.8055

Max Act: 0.3716. Min Act: -0.0000

Data Index: 983581 (C4 (Web Text))

Max Activating Token Index: 622