Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1540 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1252. Min Act: -0.0002

Data Index: 1425975 (Python Code)

Max Activating Token Index: 318

Text #1

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1274. Min Act: -0.0002

Data Index: 1600683 (Python Code)

Max Activating Token Index: 424

Text #2

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1240. Min Act: -0.0002

Data Index: 327831 (C4 (Web Text))

Max Activating Token Index: 994

Text #3

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1261. Min Act: -0.0002

Data Index: 224379 (C4 (Web Text))

Max Activating Token Index: 782

Text #4

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1269. Min Act: -0.0002

Data Index: 1433168 (Python Code)

Max Activating Token Index: 72

Text #5

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1249. Min Act: -0.0002

Data Index: 1256720 (C4 (Web Text))

Max Activating Token Index: 57

Text #6

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1206. Min Act: -0.0002

Data Index: 972062 (C4 (Web Text))

Max Activating Token Index: 372

Text #7

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1226. Min Act: -0.0002

Data Index: 1197859 (C4 (Web Text))

Max Activating Token Index: 336

Text #8

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1173. Min Act: -0.0002

Data Index: 1526015 (Python Code)

Max Activating Token Index: 632

Text #9

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1178. Min Act: -0.0002

Data Index: 777926 (C4 (Web Text))

Max Activating Token Index: 390

Text #10

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1173. Min Act: -0.0002

Data Index: 52513 (C4 (Web Text))

Max Activating Token Index: 292

Text #11

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1110. Min Act: -0.0002

Data Index: 1440829 (Python Code)

Max Activating Token Index: 274

Text #12

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1140. Min Act: -0.0002

Data Index: 362370 (C4 (Web Text))

Max Activating Token Index: 166

Text #13

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1108. Min Act: -0.0002

Data Index: 1028487 (C4 (Web Text))

Max Activating Token Index: 280

Text #14

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1144. Min Act: -0.0002

Data Index: 707264 (C4 (Web Text))

Max Activating Token Index: 396

Text #15

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1135. Min Act: -0.0002

Data Index: 1163418 (C4 (Web Text))

Max Activating Token Index: 516

Text #16

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1135. Min Act: -0.0002

Data Index: 1214231 (C4 (Web Text))

Max Activating Token Index: 191

Text #17

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1149. Min Act: -0.0002

Data Index: 91965 (C4 (Web Text))

Max Activating Token Index: 184

Text #18

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1119. Min Act: -0.0002

Data Index: 1267232 (C4 (Web Text))

Max Activating Token Index: 520

Text #19

Max Range: 0.1274. Min Range: -0.1274

Max Act: 0.1091. Min Act: -0.0002

Data Index: 194511 (C4 (Web Text))

Max Activating Token Index: 905