Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 728 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1536. Min Act: -0.0001

Data Index: 258996 (C4 (Web Text))

Max Activating Token Index: 599

Text #1

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1374. Min Act: -0.0001

Data Index: 1420279 (Python Code)

Max Activating Token Index: 815

Text #2

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1505. Min Act: -0.0001

Data Index: 637774 (C4 (Web Text))

Max Activating Token Index: 198

Text #3

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1322. Min Act: -0.0001

Data Index: 1412542 (Python Code)

Max Activating Token Index: 763

Text #4

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1379. Min Act: -0.0001

Data Index: 1493344 (Python Code)

Max Activating Token Index: 705

Text #5

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1389. Min Act: -0.0001

Data Index: 409281 (C4 (Web Text))

Max Activating Token Index: 618

Text #6

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1168. Min Act: -0.0001

Data Index: 1292321 (C4 (Web Text))

Max Activating Token Index: 170

Text #7

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1297. Min Act: -0.0001

Data Index: 458470 (C4 (Web Text))

Max Activating Token Index: 712

Text #8

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1338. Min Act: -0.0001

Data Index: 498297 (C4 (Web Text))

Max Activating Token Index: 188

Text #9

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1309. Min Act: -0.0001

Data Index: 1561974 (Python Code)

Max Activating Token Index: 660

Text #10

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1187. Min Act: -0.0001

Data Index: 1428665 (Python Code)

Max Activating Token Index: 848

Text #11

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1341. Min Act: -0.0001

Data Index: 1518207 (Python Code)

Max Activating Token Index: 773

Text #12

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1398. Min Act: -0.0001

Data Index: 718185 (C4 (Web Text))

Max Activating Token Index: 602

Text #13

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1168. Min Act: -0.0001

Data Index: 1570523 (Python Code)

Max Activating Token Index: 1001

Text #14

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1117. Min Act: -0.0001

Data Index: 178393 (C4 (Web Text))

Max Activating Token Index: 97

Text #15

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1295. Min Act: -0.0001

Data Index: 645950 (C4 (Web Text))

Max Activating Token Index: 546

Text #16

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1281. Min Act: -0.0001

Data Index: 900073 (C4 (Web Text))

Max Activating Token Index: 28

Text #17

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1215. Min Act: -0.0001

Data Index: 1656568 (Python Code)

Max Activating Token Index: 524

Text #18

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1139. Min Act: -0.0001

Data Index: 1541068 (Python Code)

Max Activating Token Index: 742

Text #19

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1204. Min Act: -0.0001

Data Index: 841681 (C4 (Web Text))

Max Activating Token Index: 575