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

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

Neuron 2009 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: 0.1493. Min Range: -0.1493

Max Act: 0.1377. Min Act: -0.0001

Data Index: 666821 (C4 (Web Text))

Max Activating Token Index: 903

Text #1

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1493. Min Act: -0.0001

Data Index: 1211641 (C4 (Web Text))

Max Activating Token Index: 323

Text #2

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1445. Min Act: -0.0001

Data Index: 886528 (C4 (Web Text))

Max Activating Token Index: 885

Text #3

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1299. Min Act: -0.0001

Data Index: 1170032 (C4 (Web Text))

Max Activating Token Index: 580

Text #4

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1303. Min Act: -0.0001

Data Index: 166891 (C4 (Web Text))

Max Activating Token Index: 120

Text #5

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1252. Min Act: -0.0001

Data Index: 1129519 (C4 (Web Text))

Max Activating Token Index: 763

Text #6

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.0959. Min Act: -0.0001

Data Index: 790247 (C4 (Web Text))

Max Activating Token Index: 280

Text #7

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1191. Min Act: -0.0001

Data Index: 665022 (C4 (Web Text))

Max Activating Token Index: 362

Text #8

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1232. Min Act: -0.0001

Data Index: 989087 (C4 (Web Text))

Max Activating Token Index: 707

Text #9

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1131. Min Act: -0.0001

Data Index: 325494 (C4 (Web Text))

Max Activating Token Index: 752

Text #10

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.0992. Min Act: -0.0001

Data Index: 271721 (C4 (Web Text))

Max Activating Token Index: 400

Text #11

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1159. Min Act: -0.0001

Data Index: 965315 (C4 (Web Text))

Max Activating Token Index: 621

Text #12

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1152. Min Act: -0.0001

Data Index: 264568 (C4 (Web Text))

Max Activating Token Index: 490

Text #13

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1161. Min Act: -0.0001

Data Index: 187988 (C4 (Web Text))

Max Activating Token Index: 443

Text #14

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1141. Min Act: -0.0001

Data Index: 1584436 (Python Code)

Max Activating Token Index: 365

Text #15

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1147. Min Act: -0.0001

Data Index: 1509988 (Python Code)

Max Activating Token Index: 292

Text #16

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1165. Min Act: -0.0001

Data Index: 1191063 (C4 (Web Text))

Max Activating Token Index: 115

Text #17

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1200. Min Act: -0.0001

Data Index: 514505 (C4 (Web Text))

Max Activating Token Index: 361

Text #18

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1080. Min Act: -0.0001

Data Index: 220275 (C4 (Web Text))

Max Activating Token Index: 422

Text #19

Max Range: 0.1493. Min Range: -0.1493

Max Act: 0.1156. Min Act: -0.0001

Data Index: 302340 (C4 (Web Text))

Max Activating Token Index: 998