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

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

Neuron 1550 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: 3.4932. Min Range: -3.4932

Max Act: 3.4932. Min Act: -0.1700

Data Index: 238329 (C4 (Web Text))

Max Activating Token Index: 240

Text #1

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.4932. Min Act: -0.1700

Data Index: 932478 (C4 (Web Text))

Max Activating Token Index: 876

Text #2

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.4775. Min Act: -0.1700

Data Index: 1341049 (C4 (Web Text))

Max Activating Token Index: 233

Text #3

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.4618. Min Act: -0.1700

Data Index: 1099702 (C4 (Web Text))

Max Activating Token Index: 43

Text #4

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.4148. Min Act: -0.1700

Data Index: 207568 (C4 (Web Text))

Max Activating Token Index: 553

Text #5

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3991. Min Act: -0.1700

Data Index: 925776 (C4 (Web Text))

Max Activating Token Index: 876

Text #6

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.4305. Min Act: -0.1700

Data Index: 349798 (C4 (Web Text))

Max Activating Token Index: 243

Text #7

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.4148. Min Act: -0.1700

Data Index: 406713 (C4 (Web Text))

Max Activating Token Index: 99

Text #8

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3834. Min Act: -0.1700

Data Index: 1028018 (C4 (Web Text))

Max Activating Token Index: 284

Text #9

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3991. Min Act: -0.1700

Data Index: 968130 (C4 (Web Text))

Max Activating Token Index: 63

Text #10

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3834. Min Act: -0.1700

Data Index: 599627 (C4 (Web Text))

Max Activating Token Index: 25

Text #11

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3677. Min Act: -0.1700

Data Index: 1175902 (C4 (Web Text))

Max Activating Token Index: 122

Text #12

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3677. Min Act: -0.1700

Data Index: 48390 (C4 (Web Text))

Max Activating Token Index: 709

Text #13

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3834. Min Act: -0.1700

Data Index: 32575 (C4 (Web Text))

Max Activating Token Index: 50

Text #14

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3834. Min Act: -0.1700

Data Index: 513060 (C4 (Web Text))

Max Activating Token Index: 702

Text #15

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3677. Min Act: -0.1700

Data Index: 158959 (C4 (Web Text))

Max Activating Token Index: 35

Text #16

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3677. Min Act: -0.1700

Data Index: 207397 (C4 (Web Text))

Max Activating Token Index: 179

Text #17

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3834. Min Act: -0.1700

Data Index: 160371 (C4 (Web Text))

Max Activating Token Index: 17

Text #18

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3677. Min Act: -0.1700

Data Index: 758835 (C4 (Web Text))

Max Activating Token Index: 25

Text #19

Max Range: 3.4932. Min Range: -3.4932

Max Act: 3.3520. Min Act: -0.1700

Data Index: 140973 (C4 (Web Text))

Max Activating Token Index: 571