Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 55 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.5250. Min Act: -0.1700

Data Index: 64185 (C4 (Web Text))

Max Activating Token Index: 90

Text #1

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.5250. Min Act: -0.1700

Data Index: 946878 (C4 (Web Text))

Max Activating Token Index: 66

Text #2

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.4780. Min Act: -0.1700

Data Index: 1074351 (C4 (Web Text))

Max Activating Token Index: 84

Text #3

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.4623. Min Act: -0.1700

Data Index: 844892 (C4 (Web Text))

Max Activating Token Index: 113

Text #4

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.4310. Min Act: -0.1700

Data Index: 555691 (C4 (Web Text))

Max Activating Token Index: 94

Text #5

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.4153. Min Act: -0.1700

Data Index: 1310080 (C4 (Web Text))

Max Activating Token Index: 110

Text #6

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.4153. Min Act: -0.1700

Data Index: 371135 (C4 (Web Text))

Max Activating Token Index: 82

Text #7

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.4153. Min Act: -0.1700

Data Index: 619025 (C4 (Web Text))

Max Activating Token Index: 69

Text #8

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.3839. Min Act: -0.1700

Data Index: 1023696 (C4 (Web Text))

Max Activating Token Index: 79

Text #9

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.3839. Min Act: -0.1700

Data Index: 282473 (C4 (Web Text))

Max Activating Token Index: 74

Text #10

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.3996. Min Act: -0.1700

Data Index: 1750 (C4 (Web Text))

Max Activating Token Index: 94

Text #11

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.3839. Min Act: -0.1700

Data Index: 460939 (C4 (Web Text))

Max Activating Token Index: 85

Text #12

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.3839. Min Act: -0.1700

Data Index: 1147249 (C4 (Web Text))

Max Activating Token Index: 97

Text #13

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.3839. Min Act: -0.1700

Data Index: 109347 (C4 (Web Text))

Max Activating Token Index: 116

Text #14

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.3683. Min Act: -0.1700

Data Index: 1552374 (Python Code)

Max Activating Token Index: 118

Text #15

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.3839. Min Act: -0.1700

Data Index: 610843 (C4 (Web Text))

Max Activating Token Index: 72

Text #16

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.3683. Min Act: -0.1700

Data Index: 468873 (C4 (Web Text))

Max Activating Token Index: 75

Text #17

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.3683. Min Act: -0.1700

Data Index: 265095 (C4 (Web Text))

Max Activating Token Index: 66

Text #18

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.3526. Min Act: -0.1700

Data Index: 885709 (C4 (Web Text))

Max Activating Token Index: 84

Text #19

Max Range: 3.5250. Min Range: -3.5250

Max Act: 3.3526. Min Act: -0.1700

Data Index: 825680 (C4 (Web Text))

Max Activating Token Index: 55