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

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

Neuron 3626 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.9043. Min Act: -0.0001

Data Index: 753013 (C4 (Web Text))

Max Activating Token Index: 291

Text #1

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.8719. Min Act: -0.0001

Data Index: 1464209 (Python Code)

Max Activating Token Index: 512

Text #2

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.8873. Min Act: -0.0000

Data Index: 148601 (C4 (Web Text))

Max Activating Token Index: 452

Text #3

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7983. Min Act: -0.0001

Data Index: 1239909 (C4 (Web Text))

Max Activating Token Index: 238

Text #4

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7716. Min Act: -0.0001

Data Index: 733010 (C4 (Web Text))

Max Activating Token Index: 221

Text #5

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7647. Min Act: -0.0001

Data Index: 328247 (C4 (Web Text))

Max Activating Token Index: 950

Text #6

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7600. Min Act: -0.0001

Data Index: 648426 (C4 (Web Text))

Max Activating Token Index: 135

Text #7

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7261. Min Act: -0.0000

Data Index: 1151704 (C4 (Web Text))

Max Activating Token Index: 291

Text #8

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7228. Min Act: -0.0001

Data Index: 1303985 (C4 (Web Text))

Max Activating Token Index: 772

Text #9

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7416. Min Act: -0.0000

Data Index: 157933 (C4 (Web Text))

Max Activating Token Index: 300

Text #10

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7290. Min Act: -0.0001

Data Index: 105424 (C4 (Web Text))

Max Activating Token Index: 457

Text #11

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7274. Min Act: -0.0000

Data Index: 924301 (C4 (Web Text))

Max Activating Token Index: 393

Text #12

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7453. Min Act: -0.0001

Data Index: 1575214 (Python Code)

Max Activating Token Index: 571

Text #13

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7188. Min Act: -0.0001

Data Index: 583777 (C4 (Web Text))

Max Activating Token Index: 590

Text #14

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7482. Min Act: -0.0000

Data Index: 217695 (C4 (Web Text))

Max Activating Token Index: 371

Text #15

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7409. Min Act: -0.0001

Data Index: 801504 (C4 (Web Text))

Max Activating Token Index: 123

Text #16

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7363. Min Act: -0.0000

Data Index: 23688 (C4 (Web Text))

Max Activating Token Index: 157

Text #17

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7375. Min Act: -0.0000

Data Index: 1352003 (C4 (Web Text))

Max Activating Token Index: 882

Text #18

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.7103. Min Act: -0.0001

Data Index: 209840 (C4 (Web Text))

Max Activating Token Index: 699

Text #19

Max Range: 0.9043. Min Range: -0.9043

Max Act: 0.6877. Min Act: -0.0001

Data Index: 1182095 (C4 (Web Text))

Max Activating Token Index: 837