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

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

Neuron 387 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1292. Min Act: -0.0001

Data Index: 819629 (C4 (Web Text))

Max Activating Token Index: 316

Text #1

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1289. Min Act: -0.0001

Data Index: 80636 (C4 (Web Text))

Max Activating Token Index: 326

Text #2

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1269. Min Act: -0.0001

Data Index: 97349 (C4 (Web Text))

Max Activating Token Index: 712

Text #3

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1239. Min Act: -0.0001

Data Index: 513172 (C4 (Web Text))

Max Activating Token Index: 302

Text #4

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1215. Min Act: -0.0001

Data Index: 1093383 (C4 (Web Text))

Max Activating Token Index: 718

Text #5

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1280. Min Act: -0.0001

Data Index: 994542 (C4 (Web Text))

Max Activating Token Index: 166

Text #6

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1223. Min Act: -0.0001

Data Index: 763402 (C4 (Web Text))

Max Activating Token Index: 286

Text #7

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1166. Min Act: -0.0001

Data Index: 176284 (C4 (Web Text))

Max Activating Token Index: 943

Text #8

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1159. Min Act: -0.0001

Data Index: 709198 (C4 (Web Text))

Max Activating Token Index: 495

Text #9

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1168. Min Act: -0.0001

Data Index: 449330 (C4 (Web Text))

Max Activating Token Index: 124

Text #10

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1124. Min Act: -0.0001

Data Index: 647653 (C4 (Web Text))

Max Activating Token Index: 700

Text #11

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1132. Min Act: -0.0001

Data Index: 1273883 (C4 (Web Text))

Max Activating Token Index: 935

Text #12

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1286. Min Act: -0.0001

Data Index: 1256806 (C4 (Web Text))

Max Activating Token Index: 262

Text #13

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1152. Min Act: -0.0001

Data Index: 1340929 (C4 (Web Text))

Max Activating Token Index: 115

Text #14

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1121. Min Act: -0.0001

Data Index: 193621 (C4 (Web Text))

Max Activating Token Index: 622

Text #15

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1090. Min Act: -0.0001

Data Index: 4584 (C4 (Web Text))

Max Activating Token Index: 868

Text #16

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1233. Min Act: -0.0001

Data Index: 1520164 (Python Code)

Max Activating Token Index: 917

Text #17

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1163. Min Act: -0.0001

Data Index: 782014 (C4 (Web Text))

Max Activating Token Index: 192

Text #18

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1079. Min Act: -0.0001

Data Index: 331624 (C4 (Web Text))

Max Activating Token Index: 453

Text #19

Max Range: 0.1292. Min Range: -0.1292

Max Act: 0.1081. Min Act: -0.0001

Data Index: 1192339 (C4 (Web Text))

Max Activating Token Index: 980