Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 512 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 5.2946. Min Range: -5.2946

Max Act: 5.2946. Min Act: -0.1700

Data Index: 627475 (C4 (Web Text))

Max Activating Token Index: 529

Text #1

Max Range: 5.2946. Min Range: -5.2946

Max Act: 5.2946. Min Act: -0.1700

Data Index: 1247438 (C4 (Web Text))

Max Activating Token Index: 894

Text #2

Max Range: 5.2946. Min Range: -5.2946

Max Act: 5.1383. Min Act: -0.1700

Data Index: 250549 (C4 (Web Text))

Max Activating Token Index: 815

Text #3

Max Range: 5.2946. Min Range: -5.2946

Max Act: 5.0758. Min Act: -0.1700

Data Index: 1194493 (C4 (Web Text))

Max Activating Token Index: 904

Text #4

Max Range: 5.2946. Min Range: -5.2946

Max Act: 5.1383. Min Act: -0.1700

Data Index: 800433 (C4 (Web Text))

Max Activating Token Index: 765

Text #5

Max Range: 5.2946. Min Range: -5.2946

Max Act: 5.0758. Min Act: -0.1700

Data Index: 52998 (C4 (Web Text))

Max Activating Token Index: 595

Text #6

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.9508. Min Act: -0.1700

Data Index: 926282 (C4 (Web Text))

Max Activating Token Index: 1017

Text #7

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.9196. Min Act: -0.1700

Data Index: 1241414 (C4 (Web Text))

Max Activating Token Index: 452

Text #8

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.8258. Min Act: -0.1700

Data Index: 1036241 (C4 (Web Text))

Max Activating Token Index: 756

Text #9

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.8571. Min Act: -0.1700

Data Index: 238210 (C4 (Web Text))

Max Activating Token Index: 598

Text #10

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.7633. Min Act: -0.1700

Data Index: 1330113 (C4 (Web Text))

Max Activating Token Index: 408

Text #11

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.7321. Min Act: -0.1700

Data Index: 1114732 (C4 (Web Text))

Max Activating Token Index: 868

Text #12

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.7321. Min Act: -0.1700

Data Index: 1241300 (C4 (Web Text))

Max Activating Token Index: 916

Text #13

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.7008. Min Act: -0.1700

Data Index: 1105010 (C4 (Web Text))

Max Activating Token Index: 257

Text #14

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.7321. Min Act: -0.1700

Data Index: 921002 (C4 (Web Text))

Max Activating Token Index: 164

Text #15

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.6383. Min Act: -0.1700

Data Index: 529164 (C4 (Web Text))

Max Activating Token Index: 194

Text #16

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.6383. Min Act: -0.1700

Data Index: 1114268 (C4 (Web Text))

Max Activating Token Index: 897

Text #17

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.6383. Min Act: -0.1700

Data Index: 1154757 (C4 (Web Text))

Max Activating Token Index: 681

Text #18

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.6071. Min Act: -0.1700

Data Index: 1416839 (Python Code)

Max Activating Token Index: 345

Text #19

Max Range: 5.2946. Min Range: -5.2946

Max Act: 4.6071. Min Act: -0.1700

Data Index: 885876 (C4 (Web Text))

Max Activating Token Index: 966