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

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

Neuron 3987 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.7536. Min Range: -5.7536

Max Act: 5.2463. Min Act: -0.0001

Data Index: 931610 (C4 (Web Text))

Max Activating Token Index: 476

Text #1

Max Range: 5.7536. Min Range: -5.7536

Max Act: 5.7536. Min Act: -0.0001

Data Index: 1058906 (C4 (Web Text))

Max Activating Token Index: 118

Text #2

Max Range: 5.7536. Min Range: -5.7536

Max Act: 4.1369. Min Act: -0.0001

Data Index: 239939 (C4 (Web Text))

Max Activating Token Index: 562

Text #3

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.9455. Min Act: -0.0001

Data Index: 1086478 (C4 (Web Text))

Max Activating Token Index: 298

Text #4

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.9819. Min Act: -0.0001

Data Index: 208627 (C4 (Web Text))

Max Activating Token Index: 485

Text #5

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.9481. Min Act: -0.0001

Data Index: 1083727 (C4 (Web Text))

Max Activating Token Index: 206

Text #6

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.9616. Min Act: -0.0001

Data Index: 653284 (C4 (Web Text))

Max Activating Token Index: 130

Text #7

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.7164. Min Act: -0.0001

Data Index: 544648 (C4 (Web Text))

Max Activating Token Index: 162

Text #8

Max Range: 5.7536. Min Range: -5.7536

Max Act: 4.0513. Min Act: -0.0001

Data Index: 1585351 (Python Code)

Max Activating Token Index: 736

Text #9

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.8066. Min Act: -0.0001

Data Index: 1214398 (C4 (Web Text))

Max Activating Token Index: 473

Text #10

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.7578. Min Act: -0.0001

Data Index: 708655 (C4 (Web Text))

Max Activating Token Index: 656

Text #11

Max Range: 5.7536. Min Range: -5.7536

Max Act: 4.3110. Min Act: -0.0001

Data Index: 241285 (C4 (Web Text))

Max Activating Token Index: 229

Text #12

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.6355. Min Act: -0.0001

Data Index: 1231316 (C4 (Web Text))

Max Activating Token Index: 105

Text #13

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.5951. Min Act: -0.0001

Data Index: 94866 (C4 (Web Text))

Max Activating Token Index: 841

Text #14

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.5968. Min Act: -0.0001

Data Index: 423924 (C4 (Web Text))

Max Activating Token Index: 526

Text #15

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.5738. Min Act: -0.0001

Data Index: 23695 (C4 (Web Text))

Max Activating Token Index: 10

Text #16

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.7463. Min Act: -0.0001

Data Index: 1177295 (C4 (Web Text))

Max Activating Token Index: 126

Text #17

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.1027. Min Act: -0.0001

Data Index: 298123 (C4 (Web Text))

Max Activating Token Index: 807

Text #18

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.4972. Min Act: -0.0001

Data Index: 900815 (C4 (Web Text))

Max Activating Token Index: 663

Text #19

Max Range: 5.7536. Min Range: -5.7536

Max Act: 3.1230. Min Act: -0.0001

Data Index: 812026 (C4 (Web Text))

Max Activating Token Index: 988