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

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

Neuron 134 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: 5.4946. Min Range: -5.4946

Max Act: 5.4946. Min Act: -0.1700

Data Index: 468354 (C4 (Web Text))

Max Activating Token Index: 155

Text #1

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.4946. Min Act: -0.1700

Data Index: 1331234 (C4 (Web Text))

Max Activating Token Index: 622

Text #2

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.3384. Min Act: -0.1699

Data Index: 644798 (C4 (Web Text))

Max Activating Token Index: 56

Text #3

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.3227. Min Act: -0.1700

Data Index: 1192296 (C4 (Web Text))

Max Activating Token Index: 621

Text #4

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.3227. Min Act: -0.1700

Data Index: 651445 (C4 (Web Text))

Max Activating Token Index: 55

Text #5

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.3227. Min Act: -0.1700

Data Index: 1237167 (C4 (Web Text))

Max Activating Token Index: 683

Text #6

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.2759. Min Act: -0.1700

Data Index: 1105785 (C4 (Web Text))

Max Activating Token Index: 100

Text #7

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.2602. Min Act: -0.1700

Data Index: 439277 (C4 (Web Text))

Max Activating Token Index: 207

Text #8

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.2915. Min Act: -0.1699

Data Index: 1129686 (C4 (Web Text))

Max Activating Token Index: 82

Text #9

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.2446. Min Act: -0.1700

Data Index: 269149 (C4 (Web Text))

Max Activating Token Index: 261

Text #10

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.1977. Min Act: -0.1700

Data Index: 567139 (C4 (Web Text))

Max Activating Token Index: 99

Text #11

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.2134. Min Act: -0.1700

Data Index: 318645 (C4 (Web Text))

Max Activating Token Index: 251

Text #12

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.2446. Min Act: -0.1700

Data Index: 1348566 (C4 (Web Text))

Max Activating Token Index: 950

Text #13

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.2446. Min Act: -0.1700

Data Index: 217096 (C4 (Web Text))

Max Activating Token Index: 510

Text #14

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.1977. Min Act: -0.1700

Data Index: 276755 (C4 (Web Text))

Max Activating Token Index: 285

Text #15

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.1665. Min Act: -0.1700

Data Index: 613831 (C4 (Web Text))

Max Activating Token Index: 94

Text #16

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.1821. Min Act: -0.1700

Data Index: 1048063 (C4 (Web Text))

Max Activating Token Index: 124

Text #17

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.1665. Min Act: -0.1700

Data Index: 912701 (C4 (Web Text))

Max Activating Token Index: 168

Text #18

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.1665. Min Act: -0.1700

Data Index: 73816 (C4 (Web Text))

Max Activating Token Index: 504

Text #19

Max Range: 5.4946. Min Range: -5.4946

Max Act: 5.1196. Min Act: -0.1700

Data Index: 1286326 (C4 (Web Text))

Max Activating Token Index: 703