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

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

Neuron 2244 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: 0.1750. Min Range: -0.1750

Max Act: 0.1750. Min Act: -0.0001

Data Index: 1124509 (C4 (Web Text))

Max Activating Token Index: 912

Text #1

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1528. Min Act: -0.0001

Data Index: 841869 (C4 (Web Text))

Max Activating Token Index: 906

Text #2

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1542. Min Act: -0.0001

Data Index: 932334 (C4 (Web Text))

Max Activating Token Index: 488

Text #3

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1468. Min Act: -0.0001

Data Index: 931295 (C4 (Web Text))

Max Activating Token Index: 622

Text #4

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1458. Min Act: -0.0001

Data Index: 1268320 (C4 (Web Text))

Max Activating Token Index: 958

Text #5

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1368. Min Act: -0.0001

Data Index: 1058680 (C4 (Web Text))

Max Activating Token Index: 714

Text #6

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1287. Min Act: -0.0001

Data Index: 1344112 (C4 (Web Text))

Max Activating Token Index: 236

Text #7

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1286. Min Act: -0.0001

Data Index: 635400 (C4 (Web Text))

Max Activating Token Index: 967

Text #8

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1191. Min Act: -0.0001

Data Index: 363121 (C4 (Web Text))

Max Activating Token Index: 732

Text #9

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1217. Min Act: -0.0001

Data Index: 1067934 (C4 (Web Text))

Max Activating Token Index: 844

Text #10

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1204. Min Act: -0.0001

Data Index: 1171247 (C4 (Web Text))

Max Activating Token Index: 247

Text #11

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1278. Min Act: -0.0001

Data Index: 1645047 (Python Code)

Max Activating Token Index: 1004

Text #12

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1165. Min Act: -0.0001

Data Index: 1094084 (C4 (Web Text))

Max Activating Token Index: 361

Text #13

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1133. Min Act: -0.0001

Data Index: 315679 (C4 (Web Text))

Max Activating Token Index: 106

Text #14

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1126. Min Act: -0.0001

Data Index: 393444 (C4 (Web Text))

Max Activating Token Index: 642

Text #15

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1163. Min Act: -0.0001

Data Index: 376752 (C4 (Web Text))

Max Activating Token Index: 168

Text #16

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1075. Min Act: -0.0001

Data Index: 1085208 (C4 (Web Text))

Max Activating Token Index: 601

Text #17

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1045. Min Act: -0.0001

Data Index: 535030 (C4 (Web Text))

Max Activating Token Index: 536

Text #18

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1161. Min Act: -0.0001

Data Index: 1289496 (C4 (Web Text))

Max Activating Token Index: 410

Text #19

Max Range: 0.1750. Min Range: -0.1750

Max Act: 0.1112. Min Act: -0.0001

Data Index: 840556 (C4 (Web Text))

Max Activating Token Index: 347