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

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

Neuron 1495 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1337. Min Act: -0.0001

Data Index: 246318 (C4 (Web Text))

Max Activating Token Index: 64

Text #1

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1393. Min Act: -0.0001

Data Index: 94213 (C4 (Web Text))

Max Activating Token Index: 641

Text #2

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1279. Min Act: -0.0001

Data Index: 1272072 (C4 (Web Text))

Max Activating Token Index: 326

Text #3

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1323. Min Act: -0.0001

Data Index: 517777 (C4 (Web Text))

Max Activating Token Index: 12

Text #4

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1200. Min Act: -0.0001

Data Index: 24517 (C4 (Web Text))

Max Activating Token Index: 882

Text #5

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1152. Min Act: -0.0001

Data Index: 1035229 (C4 (Web Text))

Max Activating Token Index: 989

Text #6

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1119. Min Act: -0.0001

Data Index: 314497 (C4 (Web Text))

Max Activating Token Index: 300

Text #7

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1140. Min Act: -0.0001

Data Index: 205802 (C4 (Web Text))

Max Activating Token Index: 267

Text #8

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1225. Min Act: -0.0001

Data Index: 253072 (C4 (Web Text))

Max Activating Token Index: 935

Text #9

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1271. Min Act: -0.0001

Data Index: 783305 (C4 (Web Text))

Max Activating Token Index: 275

Text #10

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1114. Min Act: -0.0001

Data Index: 1206666 (C4 (Web Text))

Max Activating Token Index: 920

Text #11

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1101. Min Act: -0.0001

Data Index: 451341 (C4 (Web Text))

Max Activating Token Index: 19

Text #12

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1203. Min Act: -0.0001

Data Index: 201019 (C4 (Web Text))

Max Activating Token Index: 247

Text #13

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1050. Min Act: -0.0001

Data Index: 102278 (C4 (Web Text))

Max Activating Token Index: 940

Text #14

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1094. Min Act: -0.0001

Data Index: 1446780 (Python Code)

Max Activating Token Index: 502

Text #15

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1125. Min Act: -0.0001

Data Index: 898163 (C4 (Web Text))

Max Activating Token Index: 43

Text #16

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.0939. Min Act: -0.0001

Data Index: 871221 (C4 (Web Text))

Max Activating Token Index: 845

Text #17

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1046. Min Act: -0.0001

Data Index: 1258529 (C4 (Web Text))

Max Activating Token Index: 680

Text #18

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.1077. Min Act: -0.0001

Data Index: 1175485 (C4 (Web Text))

Max Activating Token Index: 242

Text #19

Max Range: 0.1393. Min Range: -0.1393

Max Act: 0.0993. Min Act: -0.0001

Data Index: 794961 (C4 (Web Text))

Max Activating Token Index: 268