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

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

Neuron 1807 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: 7.4835. Min Range: -7.4835

Max Act: 7.4835. Min Act: -0.1700

Data Index: 183922 (C4 (Web Text))

Max Activating Token Index: 606

Text #1

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.8585. Min Act: -0.1700

Data Index: 178550 (C4 (Web Text))

Max Activating Token Index: 1009

Text #2

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.8272. Min Act: -0.1700

Data Index: 321991 (C4 (Web Text))

Max Activating Token Index: 918

Text #3

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.7335. Min Act: -0.1700

Data Index: 760824 (C4 (Web Text))

Max Activating Token Index: 767

Text #4

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.5147. Min Act: -0.1700

Data Index: 1055988 (C4 (Web Text))

Max Activating Token Index: 434

Text #5

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.6085. Min Act: -0.1700

Data Index: 1289728 (C4 (Web Text))

Max Activating Token Index: 658

Text #6

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.5772. Min Act: -0.1700

Data Index: 1336539 (C4 (Web Text))

Max Activating Token Index: 181

Text #7

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.5772. Min Act: -0.1700

Data Index: 446069 (C4 (Web Text))

Max Activating Token Index: 916

Text #8

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.5460. Min Act: -0.1700

Data Index: 1318672 (C4 (Web Text))

Max Activating Token Index: 757

Text #9

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.4835. Min Act: -0.1700

Data Index: 142845 (C4 (Web Text))

Max Activating Token Index: 956

Text #10

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.5147. Min Act: -0.1700

Data Index: 1117966 (C4 (Web Text))

Max Activating Token Index: 119

Text #11

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.5147. Min Act: -0.1700

Data Index: 245613 (C4 (Web Text))

Max Activating Token Index: 763

Text #12

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.5460. Min Act: -0.1700

Data Index: 588556 (C4 (Web Text))

Max Activating Token Index: 260

Text #13

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.5147. Min Act: -0.1700

Data Index: 799489 (C4 (Web Text))

Max Activating Token Index: 247

Text #14

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.4210. Min Act: -0.1700

Data Index: 810187 (C4 (Web Text))

Max Activating Token Index: 930

Text #15

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.4522. Min Act: -0.1700

Data Index: 235232 (C4 (Web Text))

Max Activating Token Index: 918

Text #16

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.3897. Min Act: -0.1700

Data Index: 794757 (C4 (Web Text))

Max Activating Token Index: 398

Text #17

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.4522. Min Act: -0.1700

Data Index: 1062398 (C4 (Web Text))

Max Activating Token Index: 666

Text #18

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.4210. Min Act: -0.1699

Data Index: 863863 (C4 (Web Text))

Max Activating Token Index: 170

Text #19

Max Range: 7.4835. Min Range: -7.4835

Max Act: 6.4210. Min Act: -0.1700

Data Index: 58992 (C4 (Web Text))

Max Activating Token Index: 611