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

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

Neuron 5559 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: 5.5801. Min Range: -5.5801

Max Act: 5.5801. Min Act: -0.0000

Data Index: 754126 (C4 (Web Text))

Max Activating Token Index: 301

Text #1

Max Range: 5.5801. Min Range: -5.5801

Max Act: 5.1273. Min Act: -0.0000

Data Index: 778928 (C4 (Web Text))

Max Activating Token Index: 113

Text #2

Max Range: 5.5801. Min Range: -5.5801

Max Act: 4.8050. Min Act: -0.0000

Data Index: 1328168 (C4 (Web Text))

Max Activating Token Index: 673

Text #3

Max Range: 5.5801. Min Range: -5.5801

Max Act: 4.2617. Min Act: -0.0000

Data Index: 775169 (C4 (Web Text))

Max Activating Token Index: 577

Text #4

Max Range: 5.5801. Min Range: -5.5801

Max Act: 4.0440. Min Act: -0.0000

Data Index: 1515368 (Python Code)

Max Activating Token Index: 882

Text #5

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.6895. Min Act: -0.0000

Data Index: 1428075 (Python Code)

Max Activating Token Index: 498

Text #6

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.7907. Min Act: -0.0000

Data Index: 897670 (C4 (Web Text))

Max Activating Token Index: 453

Text #7

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.9128. Min Act: -0.0000

Data Index: 1604279 (Python Code)

Max Activating Token Index: 157

Text #8

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.6588. Min Act: -0.0000

Data Index: 1564169 (Python Code)

Max Activating Token Index: 383

Text #9

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.3945. Min Act: -0.0000

Data Index: 815676 (C4 (Web Text))

Max Activating Token Index: 392

Text #10

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.2984. Min Act: -0.0000

Data Index: 1531646 (Python Code)

Max Activating Token Index: 914

Text #11

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.3541. Min Act: -0.0000

Data Index: 1470635 (Python Code)

Max Activating Token Index: 493

Text #12

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.4630. Min Act: -0.0000

Data Index: 1327675 (C4 (Web Text))

Max Activating Token Index: 932

Text #13

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.0945. Min Act: -0.0000

Data Index: 437154 (C4 (Web Text))

Max Activating Token Index: 664

Text #14

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.1583. Min Act: -0.0000

Data Index: 1529049 (Python Code)

Max Activating Token Index: 159

Text #15

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.1042. Min Act: -0.0000

Data Index: 1244557 (C4 (Web Text))

Max Activating Token Index: 40

Text #16

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.0099. Min Act: -0.0000

Data Index: 598860 (C4 (Web Text))

Max Activating Token Index: 574

Text #17

Max Range: 5.5801. Min Range: -5.5801

Max Act: 3.1587. Min Act: -0.0000

Data Index: 1463079 (Python Code)

Max Activating Token Index: 165

Text #18

Max Range: 5.5801. Min Range: -5.5801

Max Act: 2.9085. Min Act: -0.0000

Data Index: 1114331 (C4 (Web Text))

Max Activating Token Index: 174

Text #19

Max Range: 5.5801. Min Range: -5.5801

Max Act: 2.8732. Min Act: -0.0000

Data Index: 911923 (C4 (Web Text))

Max Activating Token Index: 373