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

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

Neuron 648 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.5363. Min Act: -0.0001

Data Index: 284164 (C4 (Web Text))

Max Activating Token Index: 81

Text #1

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4538. Min Act: -0.0001

Data Index: 444273 (C4 (Web Text))

Max Activating Token Index: 282

Text #2

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.5124. Min Act: -0.0001

Data Index: 1284811 (C4 (Web Text))

Max Activating Token Index: 84

Text #3

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4702. Min Act: -0.0001

Data Index: 560724 (C4 (Web Text))

Max Activating Token Index: 150

Text #4

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4857. Min Act: -0.0001

Data Index: 1029834 (C4 (Web Text))

Max Activating Token Index: 90

Text #5

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4786. Min Act: -0.0001

Data Index: 1109040 (C4 (Web Text))

Max Activating Token Index: 114

Text #6

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4486. Min Act: -0.0001

Data Index: 511947 (C4 (Web Text))

Max Activating Token Index: 504

Text #7

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4467. Min Act: -0.0001

Data Index: 317633 (C4 (Web Text))

Max Activating Token Index: 1007

Text #8

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4431. Min Act: -0.0001

Data Index: 864355 (C4 (Web Text))

Max Activating Token Index: 436

Text #9

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4390. Min Act: -0.0001

Data Index: 1055816 (C4 (Web Text))

Max Activating Token Index: 890

Text #10

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4522. Min Act: -0.0001

Data Index: 609143 (C4 (Web Text))

Max Activating Token Index: 332

Text #11

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4649. Min Act: -0.0001

Data Index: 813071 (C4 (Web Text))

Max Activating Token Index: 485

Text #12

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4351. Min Act: -0.0001

Data Index: 290175 (C4 (Web Text))

Max Activating Token Index: 312

Text #13

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4465. Min Act: -0.0001

Data Index: 365088 (C4 (Web Text))

Max Activating Token Index: 343

Text #14

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4601. Min Act: -0.0001

Data Index: 1180395 (C4 (Web Text))

Max Activating Token Index: 739

Text #15

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4557. Min Act: -0.0001

Data Index: 980480 (C4 (Web Text))

Max Activating Token Index: 999

Text #16

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4421. Min Act: -0.0001

Data Index: 1157202 (C4 (Web Text))

Max Activating Token Index: 986

Text #17

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4278. Min Act: -0.0001

Data Index: 230530 (C4 (Web Text))

Max Activating Token Index: 803

Text #18

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4386. Min Act: -0.0001

Data Index: 339557 (C4 (Web Text))

Max Activating Token Index: 317

Text #19

Max Range: 0.5363. Min Range: -0.5363

Max Act: 0.4356. Min Act: -0.0001

Data Index: 527360 (C4 (Web Text))

Max Activating Token Index: 197