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

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

Neuron 2511 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: 0.0522. Min Range: -0.0522

Max Act: 0.0467. Min Act: -0.0001

Data Index: 1068442 (C4 (Web Text))

Max Activating Token Index: 83

Text #1

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0522. Min Act: -0.0001

Data Index: 680110 (C4 (Web Text))

Max Activating Token Index: 164

Text #2

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0519. Min Act: -0.0001

Data Index: 506401 (C4 (Web Text))

Max Activating Token Index: 813

Text #3

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0435. Min Act: -0.0001

Data Index: 766673 (C4 (Web Text))

Max Activating Token Index: 188

Text #4

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0399. Min Act: -0.0001

Data Index: 865665 (C4 (Web Text))

Max Activating Token Index: 1017

Text #5

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0394. Min Act: -0.0001

Data Index: 564825 (C4 (Web Text))

Max Activating Token Index: 45

Text #6

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0378. Min Act: -0.0001

Data Index: 1174684 (C4 (Web Text))

Max Activating Token Index: 424

Text #7

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0355. Min Act: -0.0001

Data Index: 494235 (C4 (Web Text))

Max Activating Token Index: 714

Text #8

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0360. Min Act: -0.0001

Data Index: 238355 (C4 (Web Text))

Max Activating Token Index: 598

Text #9

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0320. Min Act: -0.0001

Data Index: 999439 (C4 (Web Text))

Max Activating Token Index: 141

Text #10

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0335. Min Act: -0.0001

Data Index: 14518 (C4 (Web Text))

Max Activating Token Index: 690

Text #11

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0349. Min Act: -0.0001

Data Index: 1469687 (Python Code)

Max Activating Token Index: 409

Text #12

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0326. Min Act: -0.0001

Data Index: 489386 (C4 (Web Text))

Max Activating Token Index: 456

Text #13

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0334. Min Act: -0.0001

Data Index: 1586552 (Python Code)

Max Activating Token Index: 713

Text #14

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0332. Min Act: -0.0001

Data Index: 395015 (C4 (Web Text))

Max Activating Token Index: 121

Text #15

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0348. Min Act: -0.0001

Data Index: 1234517 (C4 (Web Text))

Max Activating Token Index: 345

Text #16

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0364. Min Act: -0.0001

Data Index: 1323765 (C4 (Web Text))

Max Activating Token Index: 790

Text #17

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0358. Min Act: -0.0001

Data Index: 243284 (C4 (Web Text))

Max Activating Token Index: 167

Text #18

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0314. Min Act: -0.0001

Data Index: 1070786 (C4 (Web Text))

Max Activating Token Index: 859

Text #19

Max Range: 0.0522. Min Range: -0.0522

Max Act: 0.0334. Min Act: -0.0001

Data Index: 450898 (C4 (Web Text))

Max Activating Token Index: 879