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

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

Neuron 538 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0532. Min Act: -0.0002

Data Index: 1128604 (C4 (Web Text))

Max Activating Token Index: 982

Text #1

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0525. Min Act: -0.0001

Data Index: 971633 (C4 (Web Text))

Max Activating Token Index: 619

Text #2

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0517. Min Act: -0.0002

Data Index: 1300456 (C4 (Web Text))

Max Activating Token Index: 976

Text #3

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0528. Min Act: -0.0002

Data Index: 1315377 (C4 (Web Text))

Max Activating Token Index: 4

Text #4

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0504. Min Act: -0.0001

Data Index: 1369563 (Python Code)

Max Activating Token Index: 812

Text #5

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0496. Min Act: -0.0002

Data Index: 584208 (C4 (Web Text))

Max Activating Token Index: 998

Text #6

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0455. Min Act: -0.0002

Data Index: 84933 (C4 (Web Text))

Max Activating Token Index: 959

Text #7

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0480. Min Act: -0.0002

Data Index: 1271118 (C4 (Web Text))

Max Activating Token Index: 344

Text #8

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0545. Min Act: -0.0002

Data Index: 263208 (C4 (Web Text))

Max Activating Token Index: 525

Text #9

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0477. Min Act: -0.0002

Data Index: 1042190 (C4 (Web Text))

Max Activating Token Index: 518

Text #10

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0486. Min Act: -0.0001

Data Index: 1607363 (Python Code)

Max Activating Token Index: 43

Text #11

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0478. Min Act: -0.0001

Data Index: 1158964 (C4 (Web Text))

Max Activating Token Index: 11

Text #12

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0486. Min Act: -0.0001

Data Index: 1595659 (Python Code)

Max Activating Token Index: 814

Text #13

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0488. Min Act: -0.0001

Data Index: 1296589 (C4 (Web Text))

Max Activating Token Index: 287

Text #14

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0470. Min Act: -0.0002

Data Index: 107333 (C4 (Web Text))

Max Activating Token Index: 9

Text #15

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0408. Min Act: -0.0002

Data Index: 309308 (C4 (Web Text))

Max Activating Token Index: 712

Text #16

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0448. Min Act: -0.0001

Data Index: 636939 (C4 (Web Text))

Max Activating Token Index: 46

Text #17

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0465. Min Act: -0.0001

Data Index: 1144030 (C4 (Web Text))

Max Activating Token Index: 17

Text #18

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0493. Min Act: -0.0002

Data Index: 94616 (C4 (Web Text))

Max Activating Token Index: 647

Text #19

Max Range: 0.0545. Min Range: -0.0545

Max Act: 0.0459. Min Act: -0.0001

Data Index: 1558480 (Python Code)

Max Activating Token Index: 887