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

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

Neuron 398 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4673. Min Act: -0.1700

Data Index: 902866 (C4 (Web Text))

Max Activating Token Index: 800

Text #1

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4986. Min Act: -0.1700

Data Index: 475404 (C4 (Web Text))

Max Activating Token Index: 513

Text #2

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4516. Min Act: -0.1700

Data Index: 233112 (C4 (Web Text))

Max Activating Token Index: 706

Text #3

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4516. Min Act: -0.1700

Data Index: 496957 (C4 (Web Text))

Max Activating Token Index: 433

Text #4

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4673. Min Act: -0.1700

Data Index: 1022892 (C4 (Web Text))

Max Activating Token Index: 804

Text #5

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4673. Min Act: -0.1700

Data Index: 425237 (C4 (Web Text))

Max Activating Token Index: 580

Text #6

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4202. Min Act: -0.1700

Data Index: 1238031 (C4 (Web Text))

Max Activating Token Index: 938

Text #7

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4202. Min Act: -0.1700

Data Index: 1055888 (C4 (Web Text))

Max Activating Token Index: 441

Text #8

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4046. Min Act: -0.1700

Data Index: 734098 (C4 (Web Text))

Max Activating Token Index: 629

Text #9

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4359. Min Act: -0.1700

Data Index: 1350365 (C4 (Web Text))

Max Activating Token Index: 894

Text #10

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4046. Min Act: -0.1700

Data Index: 1252343 (C4 (Web Text))

Max Activating Token Index: 163

Text #11

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4202. Min Act: -0.1700

Data Index: 781462 (C4 (Web Text))

Max Activating Token Index: 639

Text #12

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.3889. Min Act: -0.1700

Data Index: 39708 (C4 (Web Text))

Max Activating Token Index: 452

Text #13

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4046. Min Act: -0.1700

Data Index: 20577 (C4 (Web Text))

Max Activating Token Index: 392

Text #14

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.3889. Min Act: -0.1700

Data Index: 1004694 (C4 (Web Text))

Max Activating Token Index: 441

Text #15

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.3889. Min Act: -0.1700

Data Index: 1336873 (C4 (Web Text))

Max Activating Token Index: 340

Text #16

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4202. Min Act: -0.1700

Data Index: 693134 (C4 (Web Text))

Max Activating Token Index: 274

Text #17

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4046. Min Act: -0.1700

Data Index: 127148 (C4 (Web Text))

Max Activating Token Index: 942

Text #18

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.3889. Min Act: -0.1700

Data Index: 685301 (C4 (Web Text))

Max Activating Token Index: 453

Text #19

Max Range: 3.4986. Min Range: -3.4986

Max Act: 3.4202. Min Act: -0.1700

Data Index: 424475 (C4 (Web Text))

Max Activating Token Index: 602