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

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

Neuron 3599 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1497. Min Range: -1.1497

Max Act: 1.1497. Min Act: -0.0001

Data Index: 1494317 (Python Code)

Max Activating Token Index: 169

Text #1

Max Range: 1.1497. Min Range: -1.1497

Max Act: 1.0218. Min Act: -0.0001

Data Index: 1455991 (Python Code)

Max Activating Token Index: 557

Text #2

Max Range: 1.1497. Min Range: -1.1497

Max Act: 1.1059. Min Act: -0.0001

Data Index: 1140723 (C4 (Web Text))

Max Activating Token Index: 329

Text #3

Max Range: 1.1497. Min Range: -1.1497

Max Act: 1.1129. Min Act: -0.0001

Data Index: 1320866 (C4 (Web Text))

Max Activating Token Index: 291

Text #4

Max Range: 1.1497. Min Range: -1.1497

Max Act: 1.0483. Min Act: -0.0001

Data Index: 1451698 (Python Code)

Max Activating Token Index: 885

Text #5

Max Range: 1.1497. Min Range: -1.1497

Max Act: 1.0060. Min Act: -0.0001

Data Index: 450897 (C4 (Web Text))

Max Activating Token Index: 290

Text #6

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.9562. Min Act: -0.0001

Data Index: 235390 (C4 (Web Text))

Max Activating Token Index: 886

Text #7

Max Range: 1.1497. Min Range: -1.1497

Max Act: 1.0015. Min Act: -0.0001

Data Index: 842791 (C4 (Web Text))

Max Activating Token Index: 847

Text #8

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.8984. Min Act: -0.0001

Data Index: 455822 (C4 (Web Text))

Max Activating Token Index: 610

Text #9

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.9564. Min Act: -0.0001

Data Index: 508962 (C4 (Web Text))

Max Activating Token Index: 937

Text #10

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.9573. Min Act: -0.0001

Data Index: 1455189 (Python Code)

Max Activating Token Index: 366

Text #11

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.8863. Min Act: -0.0001

Data Index: 675140 (C4 (Web Text))

Max Activating Token Index: 906

Text #12

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.9551. Min Act: -0.0001

Data Index: 1523525 (Python Code)

Max Activating Token Index: 51

Text #13

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.8683. Min Act: -0.0001

Data Index: 24387 (C4 (Web Text))

Max Activating Token Index: 429

Text #14

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.8787. Min Act: -0.0001

Data Index: 1213607 (C4 (Web Text))

Max Activating Token Index: 902

Text #15

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.7751. Min Act: -0.0001

Data Index: 1351644 (C4 (Web Text))

Max Activating Token Index: 761

Text #16

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.8865. Min Act: -0.0001

Data Index: 874829 (C4 (Web Text))

Max Activating Token Index: 931

Text #17

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.9461. Min Act: -0.0001

Data Index: 885111 (C4 (Web Text))

Max Activating Token Index: 739

Text #18

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.9146. Min Act: -0.0001

Data Index: 1071426 (C4 (Web Text))

Max Activating Token Index: 481

Text #19

Max Range: 1.1497. Min Range: -1.1497

Max Act: 0.9119. Min Act: -0.0001

Data Index: 1334533 (C4 (Web Text))

Max Activating Token Index: 585