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

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

Neuron 339 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 6.2761. Min Range: -6.2761

Max Act: 6.2761. Min Act: -0.0000

Data Index: 275680 (C4 (Web Text))

Max Activating Token Index: 117

Text #1

Max Range: 6.2761. Min Range: -6.2761

Max Act: 5.4215. Min Act: -0.0000

Data Index: 1248247 (C4 (Web Text))

Max Activating Token Index: 390

Text #2

Max Range: 6.2761. Min Range: -6.2761

Max Act: 4.7135. Min Act: -0.0000

Data Index: 815971 (C4 (Web Text))

Max Activating Token Index: 320

Text #3

Max Range: 6.2761. Min Range: -6.2761

Max Act: 5.1239. Min Act: -0.0000

Data Index: 623850 (C4 (Web Text))

Max Activating Token Index: 128

Text #4

Max Range: 6.2761. Min Range: -6.2761

Max Act: 4.5466. Min Act: -0.0000

Data Index: 1039617 (C4 (Web Text))

Max Activating Token Index: 847

Text #5

Max Range: 6.2761. Min Range: -6.2761

Max Act: 4.5116. Min Act: -0.0000

Data Index: 678918 (C4 (Web Text))

Max Activating Token Index: 36

Text #6

Max Range: 6.2761. Min Range: -6.2761

Max Act: 4.1412. Min Act: -0.0000

Data Index: 1148701 (C4 (Web Text))

Max Activating Token Index: 582

Text #7

Max Range: 6.2761. Min Range: -6.2761

Max Act: 4.0480. Min Act: -0.0000

Data Index: 1172384 (C4 (Web Text))

Max Activating Token Index: 707

Text #8

Max Range: 6.2761. Min Range: -6.2761

Max Act: 4.2305. Min Act: -0.0000

Data Index: 691536 (C4 (Web Text))

Max Activating Token Index: 46

Text #9

Max Range: 6.2761. Min Range: -6.2761

Max Act: 4.3849. Min Act: -0.0000

Data Index: 795020 (C4 (Web Text))

Max Activating Token Index: 103

Text #10

Max Range: 6.2761. Min Range: -6.2761

Max Act: 4.2299. Min Act: -0.0000

Data Index: 1013808 (C4 (Web Text))

Max Activating Token Index: 144

Text #11

Max Range: 6.2761. Min Range: -6.2761

Max Act: 3.8989. Min Act: -0.0000

Data Index: 20610 (C4 (Web Text))

Max Activating Token Index: 222

Text #12

Max Range: 6.2761. Min Range: -6.2761

Max Act: 3.9861. Min Act: -0.0000

Data Index: 1410193 (Python Code)

Max Activating Token Index: 280

Text #13

Max Range: 6.2761. Min Range: -6.2761

Max Act: 3.4783. Min Act: -0.0000

Data Index: 619712 (C4 (Web Text))

Max Activating Token Index: 973

Text #14

Max Range: 6.2761. Min Range: -6.2761

Max Act: 3.8977. Min Act: -0.0000

Data Index: 1163523 (C4 (Web Text))

Max Activating Token Index: 340

Text #15

Max Range: 6.2761. Min Range: -6.2761

Max Act: 3.7222. Min Act: -0.0000

Data Index: 1548076 (Python Code)

Max Activating Token Index: 571

Text #16

Max Range: 6.2761. Min Range: -6.2761

Max Act: 3.8443. Min Act: -0.0000

Data Index: 1359289 (C4 (Web Text))

Max Activating Token Index: 750

Text #17

Max Range: 6.2761. Min Range: -6.2761

Max Act: 3.9339. Min Act: -0.0000

Data Index: 600489 (C4 (Web Text))

Max Activating Token Index: 260

Text #18

Max Range: 6.2761. Min Range: -6.2761

Max Act: 3.9288. Min Act: -0.0000

Data Index: 962256 (C4 (Web Text))

Max Activating Token Index: 916

Text #19

Max Range: 6.2761. Min Range: -6.2761

Max Act: 3.5780. Min Act: -0.0000

Data Index: 611638 (C4 (Web Text))

Max Activating Token Index: 888