Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1948 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.8602. Min Act: -0.0001

Data Index: 1427727 (Python Code)

Max Activating Token Index: 625

Text #1

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.8951. Min Act: -0.0001

Data Index: 1376897 (Python Code)

Max Activating Token Index: 856

Text #2

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.9150. Min Act: -0.0001

Data Index: 379505 (C4 (Web Text))

Max Activating Token Index: 1014

Text #3

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.8342. Min Act: -0.0001

Data Index: 978928 (C4 (Web Text))

Max Activating Token Index: 783

Text #4

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.8491. Min Act: -0.0001

Data Index: 1481577 (Python Code)

Max Activating Token Index: 180

Text #5

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.7939. Min Act: -0.0001

Data Index: 1644745 (Python Code)

Max Activating Token Index: 198

Text #6

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.7384. Min Act: -0.0001

Data Index: 522247 (C4 (Web Text))

Max Activating Token Index: 572

Text #7

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.7523. Min Act: -0.0001

Data Index: 1558680 (Python Code)

Max Activating Token Index: 456

Text #8

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.7027. Min Act: -0.0001

Data Index: 429439 (C4 (Web Text))

Max Activating Token Index: 50

Text #9

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.6937. Min Act: -0.0001

Data Index: 1498854 (Python Code)

Max Activating Token Index: 679

Text #10

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.7367. Min Act: -0.0001

Data Index: 1464999 (Python Code)

Max Activating Token Index: 433

Text #11

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.7324. Min Act: -0.0001

Data Index: 1443282 (Python Code)

Max Activating Token Index: 782

Text #12

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.7526. Min Act: -0.0000

Data Index: 1302742 (C4 (Web Text))

Max Activating Token Index: 3

Text #13

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.7050. Min Act: -0.0001

Data Index: 1401997 (Python Code)

Max Activating Token Index: 503

Text #14

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.7259. Min Act: -0.0001

Data Index: 1619126 (Python Code)

Max Activating Token Index: 135

Text #15

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.7021. Min Act: -0.0001

Data Index: 418324 (C4 (Web Text))

Max Activating Token Index: 27

Text #16

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.6599. Min Act: -0.0001

Data Index: 1569268 (Python Code)

Max Activating Token Index: 277

Text #17

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.7031. Min Act: -0.0001

Data Index: 1399583 (Python Code)

Max Activating Token Index: 836

Text #18

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.6997. Min Act: -0.0001

Data Index: 1205421 (C4 (Web Text))

Max Activating Token Index: 571

Text #19

Max Range: 0.9150. Min Range: -0.9150

Max Act: 0.7382. Min Act: -0.0001

Data Index: 1370953 (Python Code)

Max Activating Token Index: 854