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

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

Neuron 3070 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: 0.1536. Min Range: -0.1536

Max Act: 0.1477. Min Act: -0.0001

Data Index: 1287842 (C4 (Web Text))

Max Activating Token Index: 335

Text #1

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1536. Min Act: -0.0001

Data Index: 884588 (C4 (Web Text))

Max Activating Token Index: 565

Text #2

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1449. Min Act: -0.0001

Data Index: 1026091 (C4 (Web Text))

Max Activating Token Index: 872

Text #3

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1454. Min Act: -0.0001

Data Index: 1088913 (C4 (Web Text))

Max Activating Token Index: 239

Text #4

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1397. Min Act: -0.0001

Data Index: 19053 (C4 (Web Text))

Max Activating Token Index: 350

Text #5

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1359. Min Act: -0.0001

Data Index: 879021 (C4 (Web Text))

Max Activating Token Index: 430

Text #6

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1336. Min Act: -0.0001

Data Index: 1249515 (C4 (Web Text))

Max Activating Token Index: 267

Text #7

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1268. Min Act: -0.0001

Data Index: 667296 (C4 (Web Text))

Max Activating Token Index: 505

Text #8

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1309. Min Act: -0.0001

Data Index: 1445958 (Python Code)

Max Activating Token Index: 212

Text #9

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1239. Min Act: -0.0001

Data Index: 1273569 (C4 (Web Text))

Max Activating Token Index: 428

Text #10

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1294. Min Act: -0.0001

Data Index: 498613 (C4 (Web Text))

Max Activating Token Index: 513

Text #11

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1340. Min Act: -0.0001

Data Index: 968524 (C4 (Web Text))

Max Activating Token Index: 648

Text #12

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1226. Min Act: -0.0001

Data Index: 127180 (C4 (Web Text))

Max Activating Token Index: 201

Text #13

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1200. Min Act: -0.0001

Data Index: 1056797 (C4 (Web Text))

Max Activating Token Index: 562

Text #14

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1181. Min Act: -0.0001

Data Index: 640267 (C4 (Web Text))

Max Activating Token Index: 580

Text #15

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1274. Min Act: -0.0001

Data Index: 1345352 (C4 (Web Text))

Max Activating Token Index: 808

Text #16

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1265. Min Act: -0.0001

Data Index: 19444 (C4 (Web Text))

Max Activating Token Index: 757

Text #17

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1179. Min Act: -0.0001

Data Index: 289894 (C4 (Web Text))

Max Activating Token Index: 920

Text #18

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1204. Min Act: -0.0001

Data Index: 781455 (C4 (Web Text))

Max Activating Token Index: 897

Text #19

Max Range: 0.1536. Min Range: -0.1536

Max Act: 0.1196. Min Act: -0.0001

Data Index: 639831 (C4 (Web Text))

Max Activating Token Index: 908