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

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

Neuron 457 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2315. Min Range: -5.2315

Max Act: 5.2315. Min Act: -0.1700

Data Index: 228696 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 5.2315. Min Range: -5.2315

Max Act: 5.2002. Min Act: -0.1700

Data Index: 1464812 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 5.2315. Min Range: -5.2315

Max Act: 5.1065. Min Act: -0.1700

Data Index: 1219482 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 5.2315. Min Range: -5.2315

Max Act: 5.1065. Min Act: -0.1700

Data Index: 1496995 (Python Code)

Max Activating Token Index: 1023

Text #4

Max Range: 5.2315. Min Range: -5.2315

Max Act: 5.0752. Min Act: -0.1700

Data Index: 1569792 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 5.2315. Min Range: -5.2315

Max Act: 5.1065. Min Act: -0.1700

Data Index: 1470990 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 5.2315. Min Range: -5.2315

Max Act: 5.0440. Min Act: -0.1700

Data Index: 1158802 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 5.2315. Min Range: -5.2315

Max Act: 5.0752. Min Act: -0.1700

Data Index: 552455 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 5.2315. Min Range: -5.2315

Max Act: 5.0127. Min Act: -0.1700

Data Index: 1357693 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 5.2315. Min Range: -5.2315

Max Act: 5.0440. Min Act: -0.1700

Data Index: 560085 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 5.2315. Min Range: -5.2315

Max Act: 5.0127. Min Act: -0.1700

Data Index: 1198978 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 5.2315. Min Range: -5.2315

Max Act: 5.0127. Min Act: -0.1700

Data Index: 1593324 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 5.2315. Min Range: -5.2315

Max Act: 4.9815. Min Act: -0.1700

Data Index: 729343 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 5.2315. Min Range: -5.2315

Max Act: 4.9190. Min Act: -0.1700

Data Index: 168895 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 5.2315. Min Range: -5.2315

Max Act: 4.9190. Min Act: -0.1700

Data Index: 1400944 (Python Code)

Max Activating Token Index: 1023

Text #15

Max Range: 5.2315. Min Range: -5.2315

Max Act: 4.8877. Min Act: -0.1700

Data Index: 67039 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 5.2315. Min Range: -5.2315

Max Act: 4.8877. Min Act: -0.1700

Data Index: 1621514 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 5.2315. Min Range: -5.2315

Max Act: 4.8252. Min Act: -0.1700

Data Index: 1462589 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 5.2315. Min Range: -5.2315

Max Act: 4.7940. Min Act: -0.1700

Data Index: 144567 (C4 (Web Text))

Max Activating Token Index: 701

Text #19

Max Range: 5.2315. Min Range: -5.2315

Max Act: 4.7940. Min Act: -0.1700

Data Index: 984136 (C4 (Web Text))

Max Activating Token Index: 1023