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

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

Neuron 1390 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.5729. Min Act: -0.1700

Data Index: 1648838 (Python Code)

Max Activating Token Index: 836

Text #1

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.4946. Min Act: -0.1700

Data Index: 1604810 (Python Code)

Max Activating Token Index: 202

Text #2

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.4632. Min Act: -0.1700

Data Index: 1571146 (Python Code)

Max Activating Token Index: 855

Text #3

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.4476. Min Act: -0.1700

Data Index: 265280 (C4 (Web Text))

Max Activating Token Index: 34

Text #4

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.4319. Min Act: -0.1700

Data Index: 1622915 (Python Code)

Max Activating Token Index: 474

Text #5

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.4476. Min Act: -0.1700

Data Index: 960654 (C4 (Web Text))

Max Activating Token Index: 727

Text #6

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.4476. Min Act: -0.1700

Data Index: 32154 (C4 (Web Text))

Max Activating Token Index: 312

Text #7

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.3848. Min Act: -0.1700

Data Index: 572335 (C4 (Web Text))

Max Activating Token Index: 714

Text #8

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.3691. Min Act: -0.1700

Data Index: 1467004 (Python Code)

Max Activating Token Index: 876

Text #9

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.3848. Min Act: -0.1700

Data Index: 506667 (C4 (Web Text))

Max Activating Token Index: 355

Text #10

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.3848. Min Act: -0.1700

Data Index: 1162109 (C4 (Web Text))

Max Activating Token Index: 968

Text #11

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.3848. Min Act: -0.1700

Data Index: 722712 (C4 (Web Text))

Max Activating Token Index: 440

Text #12

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.4632. Min Act: -0.1700

Data Index: 1440553 (Python Code)

Max Activating Token Index: 385

Text #13

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.3378. Min Act: -0.1700

Data Index: 606160 (C4 (Web Text))

Max Activating Token Index: 690

Text #14

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.3691. Min Act: -0.1700

Data Index: 1572249 (Python Code)

Max Activating Token Index: 747

Text #15

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.3848. Min Act: -0.1700

Data Index: 1462862 (Python Code)

Max Activating Token Index: 332

Text #16

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.3848. Min Act: -0.1700

Data Index: 1462750 (Python Code)

Max Activating Token Index: 210

Text #17

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.3378. Min Act: -0.1700

Data Index: 759072 (C4 (Web Text))

Max Activating Token Index: 779

Text #18

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.3691. Min Act: -0.1700

Data Index: 1159779 (C4 (Web Text))

Max Activating Token Index: 289

Text #19

Max Range: 3.5729. Min Range: -3.5729

Max Act: 3.3221. Min Act: -0.1700

Data Index: 1313889 (C4 (Web Text))

Max Activating Token Index: 453