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

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

Neuron 174 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.5152. Min Act: -0.0003

Data Index: 231825 (C4 (Web Text))

Max Activating Token Index: 958

Text #1

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4606. Min Act: -0.0003

Data Index: 750389 (C4 (Web Text))

Max Activating Token Index: 252

Text #2

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4835. Min Act: -0.0003

Data Index: 793297 (C4 (Web Text))

Max Activating Token Index: 457

Text #3

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.5049. Min Act: -0.0003

Data Index: 92877 (C4 (Web Text))

Max Activating Token Index: 742

Text #4

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4698. Min Act: -0.0003

Data Index: 879174 (C4 (Web Text))

Max Activating Token Index: 447

Text #5

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4220. Min Act: -0.0003

Data Index: 417375 (C4 (Web Text))

Max Activating Token Index: 499

Text #6

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.5324. Min Act: -0.0003

Data Index: 1023631 (C4 (Web Text))

Max Activating Token Index: 364

Text #7

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4924. Min Act: -0.0003

Data Index: 710297 (C4 (Web Text))

Max Activating Token Index: 629

Text #8

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4605. Min Act: -0.0003

Data Index: 267061 (C4 (Web Text))

Max Activating Token Index: 686

Text #9

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4122. Min Act: -0.0003

Data Index: 129286 (C4 (Web Text))

Max Activating Token Index: 979

Text #10

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4846. Min Act: -0.0003

Data Index: 879379 (C4 (Web Text))

Max Activating Token Index: 424

Text #11

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4889. Min Act: -0.0003

Data Index: 3639 (C4 (Web Text))

Max Activating Token Index: 967

Text #12

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4517. Min Act: -0.0003

Data Index: 1167445 (C4 (Web Text))

Max Activating Token Index: 730

Text #13

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4490. Min Act: -0.0003

Data Index: 1084220 (C4 (Web Text))

Max Activating Token Index: 382

Text #14

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4075. Min Act: -0.0003

Data Index: 512133 (C4 (Web Text))

Max Activating Token Index: 734

Text #15

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4650. Min Act: -0.0003

Data Index: 843594 (C4 (Web Text))

Max Activating Token Index: 773

Text #16

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.5269. Min Act: -0.0003

Data Index: 566640 (C4 (Web Text))

Max Activating Token Index: 490

Text #17

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.5294. Min Act: -0.0003

Data Index: 770446 (C4 (Web Text))

Max Activating Token Index: 650

Text #18

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.4477. Min Act: -0.0003

Data Index: 314399 (C4 (Web Text))

Max Activating Token Index: 591

Text #19

Max Range: 0.5324. Min Range: -0.5324

Max Act: 0.3134. Min Act: -0.0003

Data Index: 1044617 (C4 (Web Text))

Max Activating Token Index: 514