Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 615 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0601. Min Act: -0.0003

Data Index: 619338 (C4 (Web Text))

Max Activating Token Index: 182

Text #1

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0504. Min Act: -0.0003

Data Index: 1448990 (Python Code)

Max Activating Token Index: 911

Text #2

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0545. Min Act: -0.0003

Data Index: 464835 (C4 (Web Text))

Max Activating Token Index: 750

Text #3

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0542. Min Act: -0.0002

Data Index: 318578 (C4 (Web Text))

Max Activating Token Index: 425

Text #4

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0531. Min Act: -0.0003

Data Index: 1433922 (Python Code)

Max Activating Token Index: 336

Text #5

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0539. Min Act: -0.0002

Data Index: 74629 (C4 (Web Text))

Max Activating Token Index: 199

Text #6

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0519. Min Act: -0.0003

Data Index: 592586 (C4 (Web Text))

Max Activating Token Index: 575

Text #7

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0539. Min Act: -0.0003

Data Index: 1195116 (C4 (Web Text))

Max Activating Token Index: 589

Text #8

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0538. Min Act: -0.0003

Data Index: 207858 (C4 (Web Text))

Max Activating Token Index: 993

Text #9

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0546. Min Act: -0.0003

Data Index: 1542863 (Python Code)

Max Activating Token Index: 271

Text #10

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0535. Min Act: -0.0003

Data Index: 990173 (C4 (Web Text))

Max Activating Token Index: 381

Text #11

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0561. Min Act: -0.0003

Data Index: 666131 (C4 (Web Text))

Max Activating Token Index: 999

Text #12

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0499. Min Act: -0.0003

Data Index: 1538288 (Python Code)

Max Activating Token Index: 331

Text #13

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0531. Min Act: -0.0003

Data Index: 774448 (C4 (Web Text))

Max Activating Token Index: 943

Text #14

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0515. Min Act: -0.0003

Data Index: 1429773 (Python Code)

Max Activating Token Index: 172

Text #15

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0485. Min Act: -0.0003

Data Index: 566910 (C4 (Web Text))

Max Activating Token Index: 334

Text #16

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0500. Min Act: -0.0003

Data Index: 428662 (C4 (Web Text))

Max Activating Token Index: 942

Text #17

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0516. Min Act: -0.0003

Data Index: 1637701 (Python Code)

Max Activating Token Index: 251

Text #18

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0511. Min Act: -0.0003

Data Index: 541159 (C4 (Web Text))

Max Activating Token Index: 627

Text #19

Max Range: 0.0601. Min Range: -0.0601

Max Act: 0.0510. Min Act: -0.0003

Data Index: 1344906 (C4 (Web Text))

Max Activating Token Index: 109