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

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

Neuron 841 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: 5.6459. Min Range: -5.6459

Max Act: 5.6459. Min Act: -0.1699

Data Index: 424945 (C4 (Web Text))

Max Activating Token Index: 494

Text #1

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.5834. Min Act: -0.1699

Data Index: 66625 (C4 (Web Text))

Max Activating Token Index: 616

Text #2

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.3021. Min Act: -0.1699

Data Index: 1274068 (C4 (Web Text))

Max Activating Token Index: 821

Text #3

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.5209. Min Act: -0.1700

Data Index: 1401658 (Python Code)

Max Activating Token Index: 372

Text #4

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.3021. Min Act: -0.1700

Data Index: 597330 (C4 (Web Text))

Max Activating Token Index: 448

Text #5

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.3646. Min Act: -0.1698

Data Index: 374573 (C4 (Web Text))

Max Activating Token Index: 371

Text #6

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.4271. Min Act: -0.1700

Data Index: 679398 (C4 (Web Text))

Max Activating Token Index: 804

Text #7

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.4271. Min Act: -0.1700

Data Index: 1415460 (Python Code)

Max Activating Token Index: 613

Text #8

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.4584. Min Act: -0.1700

Data Index: 1605536 (Python Code)

Max Activating Token Index: 832

Text #9

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.4584. Min Act: -0.1700

Data Index: 1263845 (C4 (Web Text))

Max Activating Token Index: 530

Text #10

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.3959. Min Act: -0.1698

Data Index: 758063 (C4 (Web Text))

Max Activating Token Index: 494

Text #11

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.3959. Min Act: -0.1700

Data Index: 230191 (C4 (Web Text))

Max Activating Token Index: 705

Text #12

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.3959. Min Act: -0.1696

Data Index: 305947 (C4 (Web Text))

Max Activating Token Index: 809

Text #13

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.4271. Min Act: -0.1700

Data Index: 745349 (C4 (Web Text))

Max Activating Token Index: 768

Text #14

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.4271. Min Act: -0.1693

Data Index: 878647 (C4 (Web Text))

Max Activating Token Index: 642

Text #15

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.4584. Min Act: -0.1684

Data Index: 1620244 (Python Code)

Max Activating Token Index: 461

Text #16

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.3959. Min Act: -0.1699

Data Index: 1645277 (Python Code)

Max Activating Token Index: 329

Text #17

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.3959. Min Act: -0.1700

Data Index: 1493476 (Python Code)

Max Activating Token Index: 696

Text #18

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.3334. Min Act: -0.1698

Data Index: 1616438 (Python Code)

Max Activating Token Index: 439

Text #19

Max Range: 5.6459. Min Range: -5.6459

Max Act: 5.3959. Min Act: -0.1695

Data Index: 1363359 (Python Code)

Max Activating Token Index: 349