Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 2443 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.8536. Min Act: -0.0000

Data Index: 1593222 (Python Code)

Max Activating Token Index: 470

Text #1

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.7993. Min Act: -0.0000

Data Index: 1639091 (Python Code)

Max Activating Token Index: 489

Text #2

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.7363. Min Act: -0.0000

Data Index: 605823 (C4 (Web Text))

Max Activating Token Index: 329

Text #3

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.8111. Min Act: -0.0000

Data Index: 1587221 (Python Code)

Max Activating Token Index: 952

Text #4

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.7486. Min Act: -0.0000

Data Index: 1488155 (Python Code)

Max Activating Token Index: 552

Text #5

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.7618. Min Act: -0.0000

Data Index: 1371651 (Python Code)

Max Activating Token Index: 434

Text #6

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.7555. Min Act: -0.0000

Data Index: 1388490 (Python Code)

Max Activating Token Index: 372

Text #7

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.7334. Min Act: -0.0000

Data Index: 1360032 (Python Code)

Max Activating Token Index: 893

Text #8

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.7622. Min Act: -0.0000

Data Index: 1562349 (Python Code)

Max Activating Token Index: 600

Text #9

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.6895. Min Act: -0.0000

Data Index: 1449933 (Python Code)

Max Activating Token Index: 856

Text #10

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.6948. Min Act: -0.0000

Data Index: 1562047 (Python Code)

Max Activating Token Index: 968

Text #11

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.6267. Min Act: -0.0000

Data Index: 1412698 (Python Code)

Max Activating Token Index: 269

Text #12

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.6285. Min Act: -0.0000

Data Index: 1462955 (Python Code)

Max Activating Token Index: 711

Text #13

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.6011. Min Act: -0.0000

Data Index: 1469044 (Python Code)

Max Activating Token Index: 435

Text #14

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.6848. Min Act: -0.0000

Data Index: 1443723 (Python Code)

Max Activating Token Index: 460

Text #15

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.6674. Min Act: -0.0000

Data Index: 1589218 (Python Code)

Max Activating Token Index: 613

Text #16

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.6583. Min Act: -0.0000

Data Index: 1462524 (Python Code)

Max Activating Token Index: 492

Text #17

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.5460. Min Act: -0.0000

Data Index: 398460 (C4 (Web Text))

Max Activating Token Index: 192

Text #18

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.5472. Min Act: -0.0000

Data Index: 830940 (C4 (Web Text))

Max Activating Token Index: 613

Text #19

Max Range: 0.8536. Min Range: -0.8536

Max Act: 0.6391. Min Act: -0.0000

Data Index: 1544287 (Python Code)

Max Activating Token Index: 456