Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3899 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0435. Min Act: -0.0001

Data Index: 145055 (C4 (Web Text))

Max Activating Token Index: 28

Text #1

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0365. Min Act: -0.0001

Data Index: 1136909 (C4 (Web Text))

Max Activating Token Index: 469

Text #2

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0368. Min Act: -0.0001

Data Index: 1103662 (C4 (Web Text))

Max Activating Token Index: 577

Text #3

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0348. Min Act: -0.0001

Data Index: 19406 (C4 (Web Text))

Max Activating Token Index: 445

Text #4

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0322. Min Act: -0.0001

Data Index: 1279539 (C4 (Web Text))

Max Activating Token Index: 589

Text #5

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0355. Min Act: -0.0001

Data Index: 42859 (C4 (Web Text))

Max Activating Token Index: 294

Text #6

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0273. Min Act: -0.0001

Data Index: 396591 (C4 (Web Text))

Max Activating Token Index: 404

Text #7

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0318. Min Act: -0.0001

Data Index: 511826 (C4 (Web Text))

Max Activating Token Index: 476

Text #8

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0309. Min Act: -0.0001

Data Index: 1544266 (Python Code)

Max Activating Token Index: 162

Text #9

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0304. Min Act: -0.0001

Data Index: 1198725 (C4 (Web Text))

Max Activating Token Index: 390

Text #10

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0325. Min Act: -0.0001

Data Index: 644878 (C4 (Web Text))

Max Activating Token Index: 435

Text #11

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0313. Min Act: -0.0001

Data Index: 232116 (C4 (Web Text))

Max Activating Token Index: 424

Text #12

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0265. Min Act: -0.0001

Data Index: 334093 (C4 (Web Text))

Max Activating Token Index: 998

Text #13

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0305. Min Act: -0.0001

Data Index: 250280 (C4 (Web Text))

Max Activating Token Index: 541

Text #14

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0289. Min Act: -0.0001

Data Index: 1168074 (C4 (Web Text))

Max Activating Token Index: 2

Text #15

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0289. Min Act: -0.0001

Data Index: 1557143 (Python Code)

Max Activating Token Index: 2

Text #16

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0289. Min Act: -0.0001

Data Index: 1502975 (Python Code)

Max Activating Token Index: 2

Text #17

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0304. Min Act: -0.0001

Data Index: 150983 (C4 (Web Text))

Max Activating Token Index: 965

Text #18

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0271. Min Act: -0.0001

Data Index: 1272589 (C4 (Web Text))

Max Activating Token Index: 54

Text #19

Max Range: 0.0435. Min Range: -0.0435

Max Act: 0.0323. Min Act: -0.0001

Data Index: 1311007 (C4 (Web Text))

Max Activating Token Index: 517