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

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

Neuron 5106 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.6278. Min Act: -0.0000

Data Index: 880836 (C4 (Web Text))

Max Activating Token Index: 768

Text #1

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.6158. Min Act: -0.0000

Data Index: 838047 (C4 (Web Text))

Max Activating Token Index: 897

Text #2

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.5771. Min Act: -0.0000

Data Index: 764432 (C4 (Web Text))

Max Activating Token Index: 706

Text #3

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.5169. Min Act: -0.0000

Data Index: 421748 (C4 (Web Text))

Max Activating Token Index: 526

Text #4

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.5314. Min Act: -0.0000

Data Index: 370002 (C4 (Web Text))

Max Activating Token Index: 960

Text #5

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.5268. Min Act: -0.0000

Data Index: 710910 (C4 (Web Text))

Max Activating Token Index: 728

Text #6

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.5251. Min Act: -0.0000

Data Index: 290131 (C4 (Web Text))

Max Activating Token Index: 860

Text #7

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.5393. Min Act: -0.0000

Data Index: 9834 (C4 (Web Text))

Max Activating Token Index: 286

Text #8

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.5053. Min Act: -0.0000

Data Index: 700565 (C4 (Web Text))

Max Activating Token Index: 353

Text #9

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.5213. Min Act: -0.0000

Data Index: 1396453 (Python Code)

Max Activating Token Index: 353

Text #10

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.4954. Min Act: -0.0000

Data Index: 507375 (C4 (Web Text))

Max Activating Token Index: 677

Text #11

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.5254. Min Act: -0.0000

Data Index: 293253 (C4 (Web Text))

Max Activating Token Index: 459

Text #12

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.5064. Min Act: -0.0000

Data Index: 276743 (C4 (Web Text))

Max Activating Token Index: 400

Text #13

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.5105. Min Act: -0.0000

Data Index: 1293904 (C4 (Web Text))

Max Activating Token Index: 696

Text #14

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.5087. Min Act: -0.0000

Data Index: 1118700 (C4 (Web Text))

Max Activating Token Index: 713

Text #15

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.4950. Min Act: -0.0000

Data Index: 356803 (C4 (Web Text))

Max Activating Token Index: 538

Text #16

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.4928. Min Act: -0.0000

Data Index: 1505676 (Python Code)

Max Activating Token Index: 307

Text #17

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.4780. Min Act: -0.0000

Data Index: 1552531 (Python Code)

Max Activating Token Index: 302

Text #18

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.4774. Min Act: -0.0000

Data Index: 458642 (C4 (Web Text))

Max Activating Token Index: 895

Text #19

Max Range: 0.6278. Min Range: -0.6278

Max Act: 0.4898. Min Act: -0.0000

Data Index: 1108778 (C4 (Web Text))

Max Activating Token Index: 347