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

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

Neuron 1926 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.9460. Min Act: -0.1700

Data Index: 1456671 (Python Code)

Max Activating Token Index: 122

Text #1

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.6023. Min Act: -0.1700

Data Index: 1570779 (Python Code)

Max Activating Token Index: 525

Text #2

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.5085. Min Act: -0.1700

Data Index: 1434688 (Python Code)

Max Activating Token Index: 253

Text #3

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.5398. Min Act: -0.1700

Data Index: 1478224 (Python Code)

Max Activating Token Index: 556

Text #4

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.5398. Min Act: -0.1700

Data Index: 589118 (C4 (Web Text))

Max Activating Token Index: 853

Text #5

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.4148. Min Act: -0.1700

Data Index: 1372319 (Python Code)

Max Activating Token Index: 478

Text #6

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.4148. Min Act: -0.1700

Data Index: 1505116 (Python Code)

Max Activating Token Index: 803

Text #7

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.3210. Min Act: -0.1700

Data Index: 1482524 (Python Code)

Max Activating Token Index: 426

Text #8

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.4773. Min Act: -0.1700

Data Index: 1534539 (Python Code)

Max Activating Token Index: 540

Text #9

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.3835. Min Act: -0.1699

Data Index: 1361362 (Python Code)

Max Activating Token Index: 203

Text #10

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.3210. Min Act: -0.1699

Data Index: 1556081 (Python Code)

Max Activating Token Index: 283

Text #11

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.2585. Min Act: -0.1700

Data Index: 1383289 (Python Code)

Max Activating Token Index: 929

Text #12

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.3210. Min Act: -0.1700

Data Index: 1483185 (Python Code)

Max Activating Token Index: 522

Text #13

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.2898. Min Act: -0.1700

Data Index: 580101 (C4 (Web Text))

Max Activating Token Index: 102

Text #14

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.2585. Min Act: -0.1700

Data Index: 1468245 (Python Code)

Max Activating Token Index: 515

Text #15

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.2585. Min Act: -0.1700

Data Index: 1602631 (Python Code)

Max Activating Token Index: 854

Text #16

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.2898. Min Act: -0.1700

Data Index: 1397824 (Python Code)

Max Activating Token Index: 312

Text #17

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.2898. Min Act: -0.1700

Data Index: 1406327 (Python Code)

Max Activating Token Index: 297

Text #18

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.3523. Min Act: -0.1700

Data Index: 373159 (C4 (Web Text))

Max Activating Token Index: 810

Text #19

Max Range: 7.9460. Min Range: -7.9460

Max Act: 7.3523. Min Act: -0.1700

Data Index: 1438240 (Python Code)

Max Activating Token Index: 402