Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1915 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.8124. Min Act: -0.1700

Data Index: 350794 (C4 (Web Text))

Max Activating Token Index: 477

Text #1

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.6561. Min Act: -0.1700

Data Index: 444374 (C4 (Web Text))

Max Activating Token Index: 697

Text #2

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.5311. Min Act: -0.1700

Data Index: 668901 (C4 (Web Text))

Max Activating Token Index: 765

Text #3

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.5311. Min Act: -0.1700

Data Index: 518989 (C4 (Web Text))

Max Activating Token Index: 321

Text #4

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.4374. Min Act: -0.1700

Data Index: 500987 (C4 (Web Text))

Max Activating Token Index: 639

Text #5

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.4686. Min Act: -0.1700

Data Index: 1303857 (C4 (Web Text))

Max Activating Token Index: 633

Text #6

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.5311. Min Act: -0.1700

Data Index: 795811 (C4 (Web Text))

Max Activating Token Index: 811

Text #7

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.4686. Min Act: -0.1700

Data Index: 1187001 (C4 (Web Text))

Max Activating Token Index: 750

Text #8

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.4374. Min Act: -0.1700

Data Index: 1524260 (Python Code)

Max Activating Token Index: 460

Text #9

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.4061. Min Act: -0.1700

Data Index: 470027 (C4 (Web Text))

Max Activating Token Index: 717

Text #10

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.3749. Min Act: -0.1700

Data Index: 1077809 (C4 (Web Text))

Max Activating Token Index: 432

Text #11

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.3124. Min Act: -0.1700

Data Index: 445582 (C4 (Web Text))

Max Activating Token Index: 776

Text #12

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.3124. Min Act: -0.1700

Data Index: 133858 (C4 (Web Text))

Max Activating Token Index: 308

Text #13

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.3436. Min Act: -0.1700

Data Index: 1651243 (Python Code)

Max Activating Token Index: 673

Text #14

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.2811. Min Act: -0.1700

Data Index: 809214 (C4 (Web Text))

Max Activating Token Index: 512

Text #15

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.1874. Min Act: -0.1700

Data Index: 791399 (C4 (Web Text))

Max Activating Token Index: 1014

Text #16

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.3124. Min Act: -0.1700

Data Index: 1405492 (Python Code)

Max Activating Token Index: 705

Text #17

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.2811. Min Act: -0.1700

Data Index: 1547814 (Python Code)

Max Activating Token Index: 84

Text #18

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.2811. Min Act: -0.1700

Data Index: 1511443 (Python Code)

Max Activating Token Index: 840

Text #19

Max Range: 5.8124. Min Range: -5.8124

Max Act: 5.2811. Min Act: -0.1700

Data Index: 1633743 (Python Code)

Max Activating Token Index: 783