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

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

Neuron 1710 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2326. Min Range: -3.2326

Max Act: 3.2326. Min Act: -0.1700

Data Index: 75987 (C4 (Web Text))

Max Activating Token Index: 810

Text #1

Max Range: 3.2326. Min Range: -3.2326

Max Act: 3.1382. Min Act: -0.1700

Data Index: 618856 (C4 (Web Text))

Max Activating Token Index: 132

Text #2

Max Range: 3.2326. Min Range: -3.2326

Max Act: 3.1224. Min Act: -0.1700

Data Index: 966599 (C4 (Web Text))

Max Activating Token Index: 939

Text #3

Max Range: 3.2326. Min Range: -3.2326

Max Act: 3.1224. Min Act: -0.1700

Data Index: 1268624 (C4 (Web Text))

Max Activating Token Index: 331

Text #4

Max Range: 3.2326. Min Range: -3.2326

Max Act: 3.1067. Min Act: -0.1700

Data Index: 71165 (C4 (Web Text))

Max Activating Token Index: 161

Text #5

Max Range: 3.2326. Min Range: -3.2326

Max Act: 3.0909. Min Act: -0.1700

Data Index: 1153591 (C4 (Web Text))

Max Activating Token Index: 276

Text #6

Max Range: 3.2326. Min Range: -3.2326

Max Act: 3.0594. Min Act: -0.1700

Data Index: 518489 (C4 (Web Text))

Max Activating Token Index: 160

Text #7

Max Range: 3.2326. Min Range: -3.2326

Max Act: 3.0594. Min Act: -0.1700

Data Index: 1314917 (C4 (Web Text))

Max Activating Token Index: 23

Text #8

Max Range: 3.2326. Min Range: -3.2326

Max Act: 3.0278. Min Act: -0.1700

Data Index: 1145252 (C4 (Web Text))

Max Activating Token Index: 791

Text #9

Max Range: 3.2326. Min Range: -3.2326

Max Act: 3.0436. Min Act: -0.1700

Data Index: 1218259 (C4 (Web Text))

Max Activating Token Index: 203

Text #10

Max Range: 3.2326. Min Range: -3.2326

Max Act: 3.0278. Min Act: -0.1700

Data Index: 1211278 (C4 (Web Text))

Max Activating Token Index: 440

Text #11

Max Range: 3.2326. Min Range: -3.2326

Max Act: 2.9962. Min Act: -0.1700

Data Index: 558407 (C4 (Web Text))

Max Activating Token Index: 238

Text #12

Max Range: 3.2326. Min Range: -3.2326

Max Act: 2.9803. Min Act: -0.1700

Data Index: 761860 (C4 (Web Text))

Max Activating Token Index: 486

Text #13

Max Range: 3.2326. Min Range: -3.2326

Max Act: 2.9645. Min Act: -0.1700

Data Index: 1280270 (C4 (Web Text))

Max Activating Token Index: 740

Text #14

Max Range: 3.2326. Min Range: -3.2326

Max Act: 2.9645. Min Act: -0.1700

Data Index: 783593 (C4 (Web Text))

Max Activating Token Index: 369

Text #15

Max Range: 3.2326. Min Range: -3.2326

Max Act: 2.9487. Min Act: -0.1700

Data Index: 365714 (C4 (Web Text))

Max Activating Token Index: 845

Text #16

Max Range: 3.2326. Min Range: -3.2326

Max Act: 2.9487. Min Act: -0.1700

Data Index: 859259 (C4 (Web Text))

Max Activating Token Index: 865

Text #17

Max Range: 3.2326. Min Range: -3.2326

Max Act: 2.9329. Min Act: -0.1700

Data Index: 1585228 (Python Code)

Max Activating Token Index: 73

Text #18

Max Range: 3.2326. Min Range: -3.2326

Max Act: 2.9329. Min Act: -0.1700

Data Index: 421445 (C4 (Web Text))

Max Activating Token Index: 786

Text #19

Max Range: 3.2326. Min Range: -3.2326

Max Act: 2.9170. Min Act: -0.1700

Data Index: 402553 (C4 (Web Text))

Max Activating Token Index: 63