Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 5117 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1476. Min Act: -0.0001

Data Index: 366003 (C4 (Web Text))

Max Activating Token Index: 176

Text #1

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1517. Min Act: -0.0001

Data Index: 488644 (C4 (Web Text))

Max Activating Token Index: 97

Text #2

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1521. Min Act: -0.0001

Data Index: 1307026 (C4 (Web Text))

Max Activating Token Index: 821

Text #3

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1269. Min Act: -0.0001

Data Index: 1133270 (C4 (Web Text))

Max Activating Token Index: 189

Text #4

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1304. Min Act: -0.0001

Data Index: 1532980 (Python Code)

Max Activating Token Index: 108

Text #5

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1265. Min Act: -0.0001

Data Index: 1116844 (C4 (Web Text))

Max Activating Token Index: 144

Text #6

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1314. Min Act: -0.0001

Data Index: 353315 (C4 (Web Text))

Max Activating Token Index: 76

Text #7

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1364. Min Act: -0.0001

Data Index: 1493585 (Python Code)

Max Activating Token Index: 419

Text #8

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1412. Min Act: -0.0001

Data Index: 902431 (C4 (Web Text))

Max Activating Token Index: 394

Text #9

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1289. Min Act: -0.0001

Data Index: 305603 (C4 (Web Text))

Max Activating Token Index: 844

Text #10

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1162. Min Act: -0.0001

Data Index: 267610 (C4 (Web Text))

Max Activating Token Index: 304

Text #11

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1337. Min Act: -0.0001

Data Index: 960555 (C4 (Web Text))

Max Activating Token Index: 622

Text #12

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1081. Min Act: -0.0001

Data Index: 1448606 (Python Code)

Max Activating Token Index: 553

Text #13

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1336. Min Act: -0.0001

Data Index: 1226492 (C4 (Web Text))

Max Activating Token Index: 639

Text #14

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1199. Min Act: -0.0001

Data Index: 209000 (C4 (Web Text))

Max Activating Token Index: 348

Text #15

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1236. Min Act: -0.0001

Data Index: 683343 (C4 (Web Text))

Max Activating Token Index: 633

Text #16

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1170. Min Act: -0.0001

Data Index: 858902 (C4 (Web Text))

Max Activating Token Index: 3

Text #17

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1170. Min Act: -0.0001

Data Index: 761934 (C4 (Web Text))

Max Activating Token Index: 3

Text #18

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1199. Min Act: -0.0001

Data Index: 1494396 (Python Code)

Max Activating Token Index: 419

Text #19

Max Range: 0.1521. Min Range: -0.1521

Max Act: 0.1141. Min Act: -0.0001

Data Index: 280758 (C4 (Web Text))

Max Activating Token Index: 281