Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1550 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.7171. Min Act: -0.0001

Data Index: 471467 (C4 (Web Text))

Max Activating Token Index: 970

Text #1

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.6714. Min Act: -0.0001

Data Index: 283798 (C4 (Web Text))

Max Activating Token Index: 513

Text #2

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.6584. Min Act: -0.0001

Data Index: 307377 (C4 (Web Text))

Max Activating Token Index: 466

Text #3

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.6304. Min Act: -0.0001

Data Index: 42839 (C4 (Web Text))

Max Activating Token Index: 789

Text #4

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5842. Min Act: -0.0001

Data Index: 34649 (C4 (Web Text))

Max Activating Token Index: 939

Text #5

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5994. Min Act: -0.0001

Data Index: 1323542 (C4 (Web Text))

Max Activating Token Index: 51

Text #6

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.6050. Min Act: -0.0001

Data Index: 1283481 (C4 (Web Text))

Max Activating Token Index: 123

Text #7

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5656. Min Act: -0.0001

Data Index: 1532386 (Python Code)

Max Activating Token Index: 629

Text #8

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5725. Min Act: -0.0001

Data Index: 435963 (C4 (Web Text))

Max Activating Token Index: 246

Text #9

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5606. Min Act: -0.0001

Data Index: 517485 (C4 (Web Text))

Max Activating Token Index: 922

Text #10

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5306. Min Act: -0.0001

Data Index: 341702 (C4 (Web Text))

Max Activating Token Index: 956

Text #11

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5446. Min Act: -0.0001

Data Index: 586989 (C4 (Web Text))

Max Activating Token Index: 439

Text #12

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5549. Min Act: -0.0001

Data Index: 448983 (C4 (Web Text))

Max Activating Token Index: 615

Text #13

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5486. Min Act: -0.0001

Data Index: 547835 (C4 (Web Text))

Max Activating Token Index: 815

Text #14

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5207. Min Act: -0.0001

Data Index: 1433799 (Python Code)

Max Activating Token Index: 1013

Text #15

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5373. Min Act: -0.0001

Data Index: 567589 (C4 (Web Text))

Max Activating Token Index: 731

Text #16

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5262. Min Act: -0.0001

Data Index: 1397206 (Python Code)

Max Activating Token Index: 196

Text #17

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5286. Min Act: -0.0001

Data Index: 1251776 (C4 (Web Text))

Max Activating Token Index: 328

Text #18

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5462. Min Act: -0.0001

Data Index: 685230 (C4 (Web Text))

Max Activating Token Index: 82

Text #19

Max Range: 0.7171. Min Range: -0.7171

Max Act: 0.5201. Min Act: -0.0001

Data Index: 218417 (C4 (Web Text))

Max Activating Token Index: 81