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

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

Neuron 628 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: 3.5238. Min Range: -3.5238

Max Act: 3.5238. Min Act: -0.1700

Data Index: 325931 (C4 (Web Text))

Max Activating Token Index: 704

Text #1

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.4454. Min Act: -0.1700

Data Index: 43540 (C4 (Web Text))

Max Activating Token Index: 853

Text #2

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.4298. Min Act: -0.1700

Data Index: 1202947 (C4 (Web Text))

Max Activating Token Index: 164

Text #3

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3984. Min Act: -0.1700

Data Index: 1031833 (C4 (Web Text))

Max Activating Token Index: 829

Text #4

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3670. Min Act: -0.1700

Data Index: 887172 (C4 (Web Text))

Max Activating Token Index: 169

Text #5

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3670. Min Act: -0.1700

Data Index: 1391824 (Python Code)

Max Activating Token Index: 751

Text #6

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3513. Min Act: -0.1700

Data Index: 252901 (C4 (Web Text))

Max Activating Token Index: 62

Text #7

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3670. Min Act: -0.1700

Data Index: 511887 (C4 (Web Text))

Max Activating Token Index: 601

Text #8

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3513. Min Act: -0.1700

Data Index: 388576 (C4 (Web Text))

Max Activating Token Index: 20

Text #9

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3199. Min Act: -0.1700

Data Index: 1118803 (C4 (Web Text))

Max Activating Token Index: 299

Text #10

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3513. Min Act: -0.1700

Data Index: 783700 (C4 (Web Text))

Max Activating Token Index: 220

Text #11

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3199. Min Act: -0.1700

Data Index: 997689 (C4 (Web Text))

Max Activating Token Index: 793

Text #12

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3199. Min Act: -0.1700

Data Index: 1332200 (C4 (Web Text))

Max Activating Token Index: 831

Text #13

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3199. Min Act: -0.1700

Data Index: 801885 (C4 (Web Text))

Max Activating Token Index: 429

Text #14

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3199. Min Act: -0.1700

Data Index: 450195 (C4 (Web Text))

Max Activating Token Index: 620

Text #15

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3042. Min Act: -0.1700

Data Index: 313176 (C4 (Web Text))

Max Activating Token Index: 158

Text #16

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.2885. Min Act: -0.1700

Data Index: 687673 (C4 (Web Text))

Max Activating Token Index: 39

Text #17

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3042. Min Act: -0.1700

Data Index: 930285 (C4 (Web Text))

Max Activating Token Index: 8

Text #18

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.3042. Min Act: -0.1700

Data Index: 1269344 (C4 (Web Text))

Max Activating Token Index: 741

Text #19

Max Range: 3.5238. Min Range: -3.5238

Max Act: 3.2885. Min Act: -0.1700

Data Index: 1099831 (C4 (Web Text))

Max Activating Token Index: 84