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

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

Neuron 1711 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.2968. Min Act: -0.1700

Data Index: 1280299 (C4 (Web Text))

Max Activating Token Index: 833

Text #1

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.2968. Min Act: -0.1700

Data Index: 319432 (C4 (Web Text))

Max Activating Token Index: 244

Text #2

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.2968. Min Act: -0.1700

Data Index: 842100 (C4 (Web Text))

Max Activating Token Index: 686

Text #3

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.2811. Min Act: -0.1700

Data Index: 646328 (C4 (Web Text))

Max Activating Token Index: 475

Text #4

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.2654. Min Act: -0.1700

Data Index: 372371 (C4 (Web Text))

Max Activating Token Index: 182

Text #5

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.2654. Min Act: -0.1700

Data Index: 1354442 (C4 (Web Text))

Max Activating Token Index: 261

Text #6

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.2497. Min Act: -0.1700

Data Index: 431841 (C4 (Web Text))

Max Activating Token Index: 91

Text #7

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.2182. Min Act: -0.1700

Data Index: 1296804 (C4 (Web Text))

Max Activating Token Index: 896

Text #8

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.1867. Min Act: -0.1700

Data Index: 1281362 (C4 (Web Text))

Max Activating Token Index: 931

Text #9

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.2025. Min Act: -0.1700

Data Index: 1163934 (C4 (Web Text))

Max Activating Token Index: 704

Text #10

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.2182. Min Act: -0.1700

Data Index: 1386896 (Python Code)

Max Activating Token Index: 800

Text #11

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.2025. Min Act: -0.1700

Data Index: 857314 (C4 (Web Text))

Max Activating Token Index: 609

Text #12

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.1867. Min Act: -0.1700

Data Index: 340606 (C4 (Web Text))

Max Activating Token Index: 254

Text #13

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.1867. Min Act: -0.1700

Data Index: 62803 (C4 (Web Text))

Max Activating Token Index: 834

Text #14

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.1395. Min Act: -0.1700

Data Index: 96719 (C4 (Web Text))

Max Activating Token Index: 766

Text #15

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.1395. Min Act: -0.1700

Data Index: 1068682 (C4 (Web Text))

Max Activating Token Index: 934

Text #16

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.1710. Min Act: -0.1700

Data Index: 932399 (C4 (Web Text))

Max Activating Token Index: 776

Text #17

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.1710. Min Act: -0.1700

Data Index: 464629 (C4 (Web Text))

Max Activating Token Index: 143

Text #18

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.1553. Min Act: -0.1700

Data Index: 1613239 (Python Code)

Max Activating Token Index: 853

Text #19

Max Range: 3.2968. Min Range: -3.2968

Max Act: 3.1395. Min Act: -0.1700

Data Index: 277410 (C4 (Web Text))

Max Activating Token Index: 863