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

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

Neuron 3615 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.4050. Min Range: -2.4050

Max Act: 2.4050. Min Act: -0.0000

Data Index: 82223 (C4 (Web Text))

Max Activating Token Index: 294

Text #1

Max Range: 2.4050. Min Range: -2.4050

Max Act: 2.3591. Min Act: -0.0000

Data Index: 1518633 (Python Code)

Max Activating Token Index: 281

Text #2

Max Range: 2.4050. Min Range: -2.4050

Max Act: 2.3377. Min Act: -0.0000

Data Index: 365377 (C4 (Web Text))

Max Activating Token Index: 230

Text #3

Max Range: 2.4050. Min Range: -2.4050

Max Act: 2.2597. Min Act: -0.0000

Data Index: 895479 (C4 (Web Text))

Max Activating Token Index: 556

Text #4

Max Range: 2.4050. Min Range: -2.4050

Max Act: 2.0560. Min Act: -0.0000

Data Index: 981087 (C4 (Web Text))

Max Activating Token Index: 207

Text #5

Max Range: 2.4050. Min Range: -2.4050

Max Act: 2.1054. Min Act: -0.0000

Data Index: 683502 (C4 (Web Text))

Max Activating Token Index: 119

Text #6

Max Range: 2.4050. Min Range: -2.4050

Max Act: 2.1211. Min Act: -0.0000

Data Index: 1570020 (Python Code)

Max Activating Token Index: 356

Text #7

Max Range: 2.4050. Min Range: -2.4050

Max Act: 2.1484. Min Act: -0.0000

Data Index: 853082 (C4 (Web Text))

Max Activating Token Index: 378

Text #8

Max Range: 2.4050. Min Range: -2.4050

Max Act: 1.9748. Min Act: -0.0000

Data Index: 52403 (C4 (Web Text))

Max Activating Token Index: 120

Text #9

Max Range: 2.4050. Min Range: -2.4050

Max Act: 2.0784. Min Act: -0.0000

Data Index: 729372 (C4 (Web Text))

Max Activating Token Index: 510

Text #10

Max Range: 2.4050. Min Range: -2.4050

Max Act: 1.8541. Min Act: -0.0000

Data Index: 1605379 (Python Code)

Max Activating Token Index: 707

Text #11

Max Range: 2.4050. Min Range: -2.4050

Max Act: 1.9328. Min Act: -0.0000

Data Index: 220353 (C4 (Web Text))

Max Activating Token Index: 556

Text #12

Max Range: 2.4050. Min Range: -2.4050

Max Act: 1.9664. Min Act: -0.0000

Data Index: 586109 (C4 (Web Text))

Max Activating Token Index: 219

Text #13

Max Range: 2.4050. Min Range: -2.4050

Max Act: 2.0859. Min Act: -0.0000

Data Index: 1206269 (C4 (Web Text))

Max Activating Token Index: 60

Text #14

Max Range: 2.4050. Min Range: -2.4050

Max Act: 1.8392. Min Act: -0.0000

Data Index: 532786 (C4 (Web Text))

Max Activating Token Index: 394

Text #15

Max Range: 2.4050. Min Range: -2.4050

Max Act: 1.9395. Min Act: -0.0000

Data Index: 1270765 (C4 (Web Text))

Max Activating Token Index: 226

Text #16

Max Range: 2.4050. Min Range: -2.4050

Max Act: 1.8514. Min Act: -0.0000

Data Index: 735058 (C4 (Web Text))

Max Activating Token Index: 46

Text #17

Max Range: 2.4050. Min Range: -2.4050

Max Act: 1.7746. Min Act: -0.0000

Data Index: 1583787 (Python Code)

Max Activating Token Index: 121

Text #18

Max Range: 2.4050. Min Range: -2.4050

Max Act: 2.0606. Min Act: -0.0000

Data Index: 1110703 (C4 (Web Text))

Max Activating Token Index: 52

Text #19

Max Range: 2.4050. Min Range: -2.4050

Max Act: 1.8826. Min Act: -0.0000

Data Index: 152484 (C4 (Web Text))

Max Activating Token Index: 52