Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2799 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.9038. Min Act: -0.0001

Data Index: 1060951 (C4 (Web Text))

Max Activating Token Index: 99

Text #1

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.9038. Min Act: -0.0001

Data Index: 1261849 (C4 (Web Text))

Max Activating Token Index: 38

Text #2

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.8730. Min Act: -0.0001

Data Index: 136017 (C4 (Web Text))

Max Activating Token Index: 1008

Text #3

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.6796. Min Act: -0.0001

Data Index: 410882 (C4 (Web Text))

Max Activating Token Index: 547

Text #4

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.5290. Min Act: -0.0001

Data Index: 281047 (C4 (Web Text))

Max Activating Token Index: 61

Text #5

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.5405. Min Act: -0.0001

Data Index: 740659 (C4 (Web Text))

Max Activating Token Index: 619

Text #6

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.6102. Min Act: -0.0001

Data Index: 674660 (C4 (Web Text))

Max Activating Token Index: 124

Text #7

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.4497. Min Act: -0.0001

Data Index: 383545 (C4 (Web Text))

Max Activating Token Index: 434

Text #8

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.4135. Min Act: -0.0001

Data Index: 1219045 (C4 (Web Text))

Max Activating Token Index: 1014

Text #9

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.5752. Min Act: -0.0001

Data Index: 1015338 (C4 (Web Text))

Max Activating Token Index: 253

Text #10

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.2919. Min Act: -0.0001

Data Index: 761041 (C4 (Web Text))

Max Activating Token Index: 337

Text #11

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.3932. Min Act: -0.0001

Data Index: 283708 (C4 (Web Text))

Max Activating Token Index: 40

Text #12

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.2480. Min Act: -0.0001

Data Index: 373791 (C4 (Web Text))

Max Activating Token Index: 166

Text #13

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.2257. Min Act: -0.0001

Data Index: 487270 (C4 (Web Text))

Max Activating Token Index: 616

Text #14

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.1829. Min Act: -0.0001

Data Index: 673034 (C4 (Web Text))

Max Activating Token Index: 619

Text #15

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.3135. Min Act: -0.0001

Data Index: 217308 (C4 (Web Text))

Max Activating Token Index: 129

Text #16

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.2422. Min Act: -0.0001

Data Index: 929424 (C4 (Web Text))

Max Activating Token Index: 510

Text #17

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.3830. Min Act: -0.0001

Data Index: 968603 (C4 (Web Text))

Max Activating Token Index: 366

Text #18

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.2091. Min Act: -0.0001

Data Index: 520536 (C4 (Web Text))

Max Activating Token Index: 323

Text #19

Max Range: 3.9038. Min Range: -3.9038

Max Act: 3.0545. Min Act: -0.0001

Data Index: 1206501 (C4 (Web Text))

Max Activating Token Index: 975