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

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

Neuron 158 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.4951. Min Act: -0.1700

Data Index: 1311499 (C4 (Web Text))

Max Activating Token Index: 331

Text #1

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.4480. Min Act: -0.1700

Data Index: 1245881 (C4 (Web Text))

Max Activating Token Index: 759

Text #2

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.4794. Min Act: -0.1700

Data Index: 576429 (C4 (Web Text))

Max Activating Token Index: 720

Text #3

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.3853. Min Act: -0.1700

Data Index: 1000687 (C4 (Web Text))

Max Activating Token Index: 786

Text #4

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.4480. Min Act: -0.1700

Data Index: 853724 (C4 (Web Text))

Max Activating Token Index: 727

Text #5

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.4167. Min Act: -0.1700

Data Index: 129700 (C4 (Web Text))

Max Activating Token Index: 574

Text #6

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.4167. Min Act: -0.1700

Data Index: 1184209 (C4 (Web Text))

Max Activating Token Index: 142

Text #7

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.3853. Min Act: -0.1700

Data Index: 1075908 (C4 (Web Text))

Max Activating Token Index: 912

Text #8

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.4010. Min Act: -0.1700

Data Index: 42010 (C4 (Web Text))

Max Activating Token Index: 894

Text #9

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.4324. Min Act: -0.1700

Data Index: 749683 (C4 (Web Text))

Max Activating Token Index: 869

Text #10

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.4010. Min Act: -0.1700

Data Index: 893121 (C4 (Web Text))

Max Activating Token Index: 926

Text #11

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.4167. Min Act: -0.1700

Data Index: 975825 (C4 (Web Text))

Max Activating Token Index: 557

Text #12

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.3853. Min Act: -0.1700

Data Index: 127742 (C4 (Web Text))

Max Activating Token Index: 748

Text #13

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.3696. Min Act: -0.1700

Data Index: 76101 (C4 (Web Text))

Max Activating Token Index: 518

Text #14

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.3853. Min Act: -0.1700

Data Index: 875918 (C4 (Web Text))

Max Activating Token Index: 566

Text #15

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.3382. Min Act: -0.1700

Data Index: 149074 (C4 (Web Text))

Max Activating Token Index: 594

Text #16

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.3382. Min Act: -0.1700

Data Index: 495196 (C4 (Web Text))

Max Activating Token Index: 620

Text #17

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.3853. Min Act: -0.1700

Data Index: 860856 (C4 (Web Text))

Max Activating Token Index: 78

Text #18

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.4010. Min Act: -0.1700

Data Index: 214408 (C4 (Web Text))

Max Activating Token Index: 685

Text #19

Max Range: 3.4951. Min Range: -3.4951

Max Act: 3.3539. Min Act: -0.1700

Data Index: 599447 (C4 (Web Text))

Max Activating Token Index: 429