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

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

Neuron 2506 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: 0.1512. Min Range: -0.1512

Max Act: 0.1512. Min Act: -0.0001

Data Index: 682117 (C4 (Web Text))

Max Activating Token Index: 758

Text #1

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1393. Min Act: -0.0000

Data Index: 1638513 (Python Code)

Max Activating Token Index: 943

Text #2

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1278. Min Act: -0.0000

Data Index: 428625 (C4 (Web Text))

Max Activating Token Index: 474

Text #3

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1385. Min Act: -0.0001

Data Index: 1022787 (C4 (Web Text))

Max Activating Token Index: 384

Text #4

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1275. Min Act: -0.0001

Data Index: 1252055 (C4 (Web Text))

Max Activating Token Index: 232

Text #5

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1276. Min Act: -0.0001

Data Index: 142482 (C4 (Web Text))

Max Activating Token Index: 473

Text #6

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1190. Min Act: -0.0000

Data Index: 1435381 (Python Code)

Max Activating Token Index: 984

Text #7

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1185. Min Act: -0.0001

Data Index: 1171243 (C4 (Web Text))

Max Activating Token Index: 302

Text #8

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1164. Min Act: -0.0001

Data Index: 780633 (C4 (Web Text))

Max Activating Token Index: 13

Text #9

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1202. Min Act: -0.0000

Data Index: 1597659 (Python Code)

Max Activating Token Index: 574

Text #10

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1190. Min Act: -0.0000

Data Index: 1491057 (Python Code)

Max Activating Token Index: 454

Text #11

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1202. Min Act: -0.0001

Data Index: 92527 (C4 (Web Text))

Max Activating Token Index: 788

Text #12

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1228. Min Act: -0.0001

Data Index: 327135 (C4 (Web Text))

Max Activating Token Index: 440

Text #13

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1223. Min Act: -0.0001

Data Index: 1537807 (Python Code)

Max Activating Token Index: 551

Text #14

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1111. Min Act: -0.0001

Data Index: 427727 (C4 (Web Text))

Max Activating Token Index: 811

Text #15

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1149. Min Act: -0.0001

Data Index: 1340896 (C4 (Web Text))

Max Activating Token Index: 888

Text #16

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1093. Min Act: -0.0001

Data Index: 1625174 (Python Code)

Max Activating Token Index: 141

Text #17

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1135. Min Act: -0.0001

Data Index: 558298 (C4 (Web Text))

Max Activating Token Index: 62

Text #18

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.0989. Min Act: -0.0001

Data Index: 1431680 (Python Code)

Max Activating Token Index: 190

Text #19

Max Range: 0.1512. Min Range: -0.1512

Max Act: 0.1045. Min Act: -0.0001

Data Index: 1240206 (C4 (Web Text))

Max Activating Token Index: 434