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

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

Neuron 820 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.8289. Min Act: -0.1700

Data Index: 1078797 (C4 (Web Text))

Max Activating Token Index: 125

Text #1

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.7976. Min Act: -0.1700

Data Index: 353550 (C4 (Web Text))

Max Activating Token Index: 66

Text #2

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.6726. Min Act: -0.1700

Data Index: 249809 (C4 (Web Text))

Max Activating Token Index: 441

Text #3

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.7351. Min Act: -0.1700

Data Index: 1344295 (C4 (Web Text))

Max Activating Token Index: 113

Text #4

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.6101. Min Act: -0.1700

Data Index: 654807 (C4 (Web Text))

Max Activating Token Index: 980

Text #5

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.5789. Min Act: -0.1700

Data Index: 54331 (C4 (Web Text))

Max Activating Token Index: 846

Text #6

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.6414. Min Act: -0.1700

Data Index: 1006045 (C4 (Web Text))

Max Activating Token Index: 436

Text #7

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.6101. Min Act: -0.1700

Data Index: 1274978 (C4 (Web Text))

Max Activating Token Index: 989

Text #8

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.5476. Min Act: -0.1700

Data Index: 261555 (C4 (Web Text))

Max Activating Token Index: 727

Text #9

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.6101. Min Act: -0.1700

Data Index: 735605 (C4 (Web Text))

Max Activating Token Index: 860

Text #10

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.4851. Min Act: -0.1700

Data Index: 486464 (C4 (Web Text))

Max Activating Token Index: 789

Text #11

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.5476. Min Act: -0.1700

Data Index: 416300 (C4 (Web Text))

Max Activating Token Index: 220

Text #12

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.5476. Min Act: -0.1700

Data Index: 67605 (C4 (Web Text))

Max Activating Token Index: 947

Text #13

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.5164. Min Act: -0.1700

Data Index: 199457 (C4 (Web Text))

Max Activating Token Index: 100

Text #14

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.5476. Min Act: -0.1700

Data Index: 301713 (C4 (Web Text))

Max Activating Token Index: 97

Text #15

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.5164. Min Act: -0.1700

Data Index: 82089 (C4 (Web Text))

Max Activating Token Index: 873

Text #16

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.5476. Min Act: -0.1700

Data Index: 617875 (C4 (Web Text))

Max Activating Token Index: 796

Text #17

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.4851. Min Act: -0.1700

Data Index: 410838 (C4 (Web Text))

Max Activating Token Index: 1008

Text #18

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.4851. Min Act: -0.1700

Data Index: 760449 (C4 (Web Text))

Max Activating Token Index: 702

Text #19

Max Range: 5.8289. Min Range: -5.8289

Max Act: 5.5476. Min Act: -0.1700

Data Index: 163425 (C4 (Web Text))

Max Activating Token Index: 909