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

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

Neuron 806 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.4191. Min Act: -0.1700

Data Index: 1114079 (C4 (Web Text))

Max Activating Token Index: 70

Text #1

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.3720. Min Act: -0.1700

Data Index: 1036774 (C4 (Web Text))

Max Activating Token Index: 98

Text #2

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.3563. Min Act: -0.1700

Data Index: 1116591 (C4 (Web Text))

Max Activating Token Index: 35

Text #3

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.2150. Min Act: -0.1700

Data Index: 531432 (C4 (Web Text))

Max Activating Token Index: 885

Text #4

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.2307. Min Act: -0.1700

Data Index: 62402 (C4 (Web Text))

Max Activating Token Index: 665

Text #5

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.2150. Min Act: -0.1700

Data Index: 487091 (C4 (Web Text))

Max Activating Token Index: 889

Text #6

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1992. Min Act: -0.1700

Data Index: 456983 (C4 (Web Text))

Max Activating Token Index: 783

Text #7

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1992. Min Act: -0.1700

Data Index: 831911 (C4 (Web Text))

Max Activating Token Index: 222

Text #8

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1835. Min Act: -0.1700

Data Index: 725858 (C4 (Web Text))

Max Activating Token Index: 549

Text #9

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1992. Min Act: -0.1700

Data Index: 899007 (C4 (Web Text))

Max Activating Token Index: 130

Text #10

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1992. Min Act: -0.1700

Data Index: 997561 (C4 (Web Text))

Max Activating Token Index: 861

Text #11

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1677. Min Act: -0.1700

Data Index: 91151 (C4 (Web Text))

Max Activating Token Index: 463

Text #12

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1835. Min Act: -0.1700

Data Index: 166389 (C4 (Web Text))

Max Activating Token Index: 13

Text #13

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1677. Min Act: -0.1700

Data Index: 674536 (C4 (Web Text))

Max Activating Token Index: 821

Text #14

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1520. Min Act: -0.1700

Data Index: 280404 (C4 (Web Text))

Max Activating Token Index: 942

Text #15

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1520. Min Act: -0.1700

Data Index: 715477 (C4 (Web Text))

Max Activating Token Index: 917

Text #16

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1520. Min Act: -0.1700

Data Index: 109739 (C4 (Web Text))

Max Activating Token Index: 34

Text #17

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1520. Min Act: -0.1700

Data Index: 1264720 (C4 (Web Text))

Max Activating Token Index: 201

Text #18

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1520. Min Act: -0.1700

Data Index: 423819 (C4 (Web Text))

Max Activating Token Index: 700

Text #19

Max Range: 3.4191. Min Range: -3.4191

Max Act: 3.1362. Min Act: -0.1700

Data Index: 234309 (C4 (Web Text))

Max Activating Token Index: 814