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

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

Neuron 177 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: 3.5718. Min Range: -3.5718

Max Act: 3.5718. Min Act: -0.1700

Data Index: 824462 (C4 (Web Text))

Max Activating Token Index: 33

Text #1

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.5091. Min Act: -0.1700

Data Index: 663189 (C4 (Web Text))

Max Activating Token Index: 211

Text #2

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.4464. Min Act: -0.1700

Data Index: 383982 (C4 (Web Text))

Max Activating Token Index: 760

Text #3

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.4151. Min Act: -0.1700

Data Index: 674758 (C4 (Web Text))

Max Activating Token Index: 650

Text #4

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.3837. Min Act: -0.1700

Data Index: 402318 (C4 (Web Text))

Max Activating Token Index: 822

Text #5

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.4151. Min Act: -0.1700

Data Index: 39962 (C4 (Web Text))

Max Activating Token Index: 810

Text #6

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.3523. Min Act: -0.1700

Data Index: 818752 (C4 (Web Text))

Max Activating Token Index: 185

Text #7

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.3523. Min Act: -0.1700

Data Index: 560945 (C4 (Web Text))

Max Activating Token Index: 846

Text #8

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.3523. Min Act: -0.1700

Data Index: 1089137 (C4 (Web Text))

Max Activating Token Index: 276

Text #9

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.3052. Min Act: -0.1700

Data Index: 383723 (C4 (Web Text))

Max Activating Token Index: 829

Text #10

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.3366. Min Act: -0.1700

Data Index: 953808 (C4 (Web Text))

Max Activating Token Index: 599

Text #11

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.3366. Min Act: -0.1700

Data Index: 1495392 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.3366. Min Act: -0.1700

Data Index: 335766 (C4 (Web Text))

Max Activating Token Index: 146

Text #13

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.3052. Min Act: -0.1700

Data Index: 759130 (C4 (Web Text))

Max Activating Token Index: 724

Text #14

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.3052. Min Act: -0.1700

Data Index: 561163 (C4 (Web Text))

Max Activating Token Index: 1017

Text #15

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.2895. Min Act: -0.1700

Data Index: 815610 (C4 (Web Text))

Max Activating Token Index: 680

Text #16

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.2895. Min Act: -0.1700

Data Index: 646959 (C4 (Web Text))

Max Activating Token Index: 345

Text #17

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.3052. Min Act: -0.1700

Data Index: 111184 (C4 (Web Text))

Max Activating Token Index: 911

Text #18

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.3209. Min Act: -0.1700

Data Index: 870606 (C4 (Web Text))

Max Activating Token Index: 1001

Text #19

Max Range: 3.5718. Min Range: -3.5718

Max Act: 3.2738. Min Act: -0.1700

Data Index: 112587 (C4 (Web Text))

Max Activating Token Index: 950