Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

Dataset: The Pile

Neuron 1704 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l-pile')

Text #0

Max Range: 1.3405. Min Range: -1.3405

Max Act: 1.3405. Min Act: -0.0000

Data Index: 1876558 (The Pile)

Max Activating Token Index: 109

Text #1

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.9063. Min Act: -0.0000

Data Index: 392121 (The Pile)

Max Activating Token Index: 267

Text #2

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.7960. Min Act: -0.0000

Data Index: 99310 (The Pile)

Max Activating Token Index: 178

Text #3

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.7597. Min Act: -0.0000

Data Index: 874308 (The Pile)

Max Activating Token Index: 50

Text #4

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.7405. Min Act: -0.0000

Data Index: 435368 (The Pile)

Max Activating Token Index: 942

Text #5

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.7185. Min Act: -0.0000

Data Index: 252862 (The Pile)

Max Activating Token Index: 689

Text #6

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.6862. Min Act: -0.0001

Data Index: 692818 (The Pile)

Max Activating Token Index: 560

Text #7

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.6687. Min Act: -0.0000

Data Index: 1587208 (The Pile)

Max Activating Token Index: 204

Text #8

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.6732. Min Act: -0.0000

Data Index: 1869718 (The Pile)

Max Activating Token Index: 212

Text #9

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.6676. Min Act: -0.0000

Data Index: 1806148 (The Pile)

Max Activating Token Index: 743

Text #10

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.6795. Min Act: -0.0000

Data Index: 130707 (The Pile)

Max Activating Token Index: 283

Text #11

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.6280. Min Act: -0.0000

Data Index: 1244001 (The Pile)

Max Activating Token Index: 741

Text #12

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.6217. Min Act: -0.0001

Data Index: 642776 (The Pile)

Max Activating Token Index: 882

Text #13

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.6102. Min Act: -0.0000

Data Index: 224013 (The Pile)

Max Activating Token Index: 686

Text #14

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.5899. Min Act: -0.0000

Data Index: 256600 (The Pile)

Max Activating Token Index: 419

Text #15

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.5713. Min Act: -0.0000

Data Index: 252040 (The Pile)

Max Activating Token Index: 825

Text #16

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.6186. Min Act: -0.0000

Data Index: 1173306 (The Pile)

Max Activating Token Index: 128

Text #17

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.6008. Min Act: -0.0000

Data Index: 967914 (The Pile)

Max Activating Token Index: 776

Text #18

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.6164. Min Act: -0.0001

Data Index: 1505920 (The Pile)

Max Activating Token Index: 196

Text #19

Max Range: 1.3405. Min Range: -1.3405

Max Act: 0.5708. Min Act: -0.0000

Data Index: 1564442 (The Pile)

Max Activating Token Index: 472