Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1572 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.9523. Min Act: -0.1700

Data Index: 1511941 (Python Code)

Max Activating Token Index: 105

Text #1

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.9211. Min Act: -0.1699

Data Index: 1650623 (Python Code)

Max Activating Token Index: 758

Text #2

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.8898. Min Act: -0.1700

Data Index: 1502025 (Python Code)

Max Activating Token Index: 620

Text #3

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.9211. Min Act: -0.1698

Data Index: 1453229 (Python Code)

Max Activating Token Index: 797

Text #4

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.8586. Min Act: -0.1688

Data Index: 1403959 (Python Code)

Max Activating Token Index: 502

Text #5

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.9211. Min Act: -0.1700

Data Index: 1425912 (Python Code)

Max Activating Token Index: 984

Text #6

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.8586. Min Act: -0.1700

Data Index: 1568879 (Python Code)

Max Activating Token Index: 449

Text #7

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.8273. Min Act: -0.1700

Data Index: 1652151 (Python Code)

Max Activating Token Index: 652

Text #8

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.8898. Min Act: -0.1700

Data Index: 1608879 (Python Code)

Max Activating Token Index: 696

Text #9

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.7961. Min Act: -0.1699

Data Index: 1442824 (Python Code)

Max Activating Token Index: 403

Text #10

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.8273. Min Act: -0.1699

Data Index: 1371857 (Python Code)

Max Activating Token Index: 590

Text #11

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.8273. Min Act: -0.1700

Data Index: 1494779 (Python Code)

Max Activating Token Index: 280

Text #12

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.7961. Min Act: -0.1700

Data Index: 1610617 (Python Code)

Max Activating Token Index: 52

Text #13

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.7961. Min Act: -0.1700

Data Index: 1609777 (Python Code)

Max Activating Token Index: 121

Text #14

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.7648. Min Act: -0.1700

Data Index: 1617781 (Python Code)

Max Activating Token Index: 723

Text #15

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.7961. Min Act: -0.1700

Data Index: 1637525 (Python Code)

Max Activating Token Index: 325

Text #16

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.7961. Min Act: -0.1700

Data Index: 1638772 (Python Code)

Max Activating Token Index: 142

Text #17

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.7961. Min Act: -0.1700

Data Index: 1444701 (Python Code)

Max Activating Token Index: 571

Text #18

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.7961. Min Act: -0.1700

Data Index: 1645236 (Python Code)

Max Activating Token Index: 92

Text #19

Max Range: 4.9523. Min Range: -4.9523

Max Act: 4.7648. Min Act: -0.1700

Data Index: 1477015 (Python Code)

Max Activating Token Index: 987