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

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

Neuron 575 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.0212. Min Range: -4.0212

Max Act: 4.0212. Min Act: -0.1700

Data Index: 1031483 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.8961. Min Act: -0.1700

Data Index: 276540 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.8961. Min Act: -0.1700

Data Index: 1472656 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.9274. Min Act: -0.1700

Data Index: 931411 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.8961. Min Act: -0.1700

Data Index: 909457 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.8961. Min Act: -0.1700

Data Index: 913778 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.8335. Min Act: -0.1700

Data Index: 1331687 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.8023. Min Act: -0.1700

Data Index: 1373682 (Python Code)

Max Activating Token Index: 1023

Text #8

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.8335. Min Act: -0.1700

Data Index: 1023317 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.7710. Min Act: -0.1699

Data Index: 1195661 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.7710. Min Act: -0.1700

Data Index: 419340 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.8023. Min Act: -0.1700

Data Index: 91540 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.7710. Min Act: -0.1700

Data Index: 300644 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.7397. Min Act: -0.1700

Data Index: 771167 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.7710. Min Act: -0.1700

Data Index: 806648 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.7710. Min Act: -0.1700

Data Index: 459847 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.7397. Min Act: -0.1700

Data Index: 152473 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.7710. Min Act: -0.1700

Data Index: 1202008 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.6458. Min Act: -0.1700

Data Index: 1574835 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 4.0212. Min Range: -4.0212

Max Act: 3.6771. Min Act: -0.1700

Data Index: 1446018 (Python Code)

Max Activating Token Index: 1023