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

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

Neuron 1535 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1882. Min Act: -0.0001

Data Index: 852590 (C4 (Web Text))

Max Activating Token Index: 211

Text #1

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1514. Min Act: -0.0001

Data Index: 1074597 (C4 (Web Text))

Max Activating Token Index: 55

Text #2

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1496. Min Act: -0.0001

Data Index: 1655550 (Python Code)

Max Activating Token Index: 78

Text #3

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1574. Min Act: -0.0001

Data Index: 1551480 (Python Code)

Max Activating Token Index: 988

Text #4

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1488. Min Act: -0.0001

Data Index: 1512243 (Python Code)

Max Activating Token Index: 333

Text #5

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1338. Min Act: -0.0001

Data Index: 578946 (C4 (Web Text))

Max Activating Token Index: 274

Text #6

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1342. Min Act: -0.0001

Data Index: 449439 (C4 (Web Text))

Max Activating Token Index: 940

Text #7

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1286. Min Act: -0.0001

Data Index: 320883 (C4 (Web Text))

Max Activating Token Index: 205

Text #8

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1279. Min Act: -0.0001

Data Index: 14422 (C4 (Web Text))

Max Activating Token Index: 138

Text #9

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1228. Min Act: -0.0001

Data Index: 172266 (C4 (Web Text))

Max Activating Token Index: 980

Text #10

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1454. Min Act: -0.0001

Data Index: 1044491 (C4 (Web Text))

Max Activating Token Index: 913

Text #11

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1314. Min Act: -0.0001

Data Index: 1476441 (Python Code)

Max Activating Token Index: 582

Text #12

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1227. Min Act: -0.0001

Data Index: 1516891 (Python Code)

Max Activating Token Index: 623

Text #13

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1259. Min Act: -0.0001

Data Index: 729786 (C4 (Web Text))

Max Activating Token Index: 833

Text #14

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1302. Min Act: -0.0001

Data Index: 677232 (C4 (Web Text))

Max Activating Token Index: 300

Text #15

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1242. Min Act: -0.0001

Data Index: 678161 (C4 (Web Text))

Max Activating Token Index: 454

Text #16

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1175. Min Act: -0.0001

Data Index: 1185256 (C4 (Web Text))

Max Activating Token Index: 195

Text #17

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1050. Min Act: -0.0001

Data Index: 354535 (C4 (Web Text))

Max Activating Token Index: 513

Text #18

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1243. Min Act: -0.0001

Data Index: 1136251 (C4 (Web Text))

Max Activating Token Index: 220

Text #19

Max Range: 0.1882. Min Range: -0.1882

Max Act: 0.1329. Min Act: -0.0001

Data Index: 1182311 (C4 (Web Text))

Max Activating Token Index: 867