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

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

Neuron 1175 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: 3.3095. Min Range: -3.3095

Max Act: 3.3095. Min Act: -0.1700

Data Index: 1159418 (C4 (Web Text))

Max Activating Token Index: 618

Text #1

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.3095. Min Act: -0.1700

Data Index: 111271 (C4 (Web Text))

Max Activating Token Index: 715

Text #2

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.2938. Min Act: -0.1700

Data Index: 626244 (C4 (Web Text))

Max Activating Token Index: 39

Text #3

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.2623. Min Act: -0.1700

Data Index: 278279 (C4 (Web Text))

Max Activating Token Index: 5

Text #4

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.1837. Min Act: -0.1700

Data Index: 1189689 (C4 (Web Text))

Max Activating Token Index: 92

Text #5

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.1837. Min Act: -0.1700

Data Index: 1298728 (C4 (Web Text))

Max Activating Token Index: 684

Text #6

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.1994. Min Act: -0.1700

Data Index: 456311 (C4 (Web Text))

Max Activating Token Index: 753

Text #7

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.1837. Min Act: -0.1700

Data Index: 1303165 (C4 (Web Text))

Max Activating Token Index: 318

Text #8

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.1680. Min Act: -0.1700

Data Index: 308905 (C4 (Web Text))

Max Activating Token Index: 560

Text #9

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.1680. Min Act: -0.1700

Data Index: 961852 (C4 (Web Text))

Max Activating Token Index: 717

Text #10

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.0892. Min Act: -0.1700

Data Index: 499491 (C4 (Web Text))

Max Activating Token Index: 771

Text #11

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.1207. Min Act: -0.1700

Data Index: 600438 (C4 (Web Text))

Max Activating Token Index: 277

Text #12

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.0734. Min Act: -0.1700

Data Index: 1073484 (C4 (Web Text))

Max Activating Token Index: 659

Text #13

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.0576. Min Act: -0.1700

Data Index: 376554 (C4 (Web Text))

Max Activating Token Index: 205

Text #14

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.0734. Min Act: -0.1700

Data Index: 310090 (C4 (Web Text))

Max Activating Token Index: 599

Text #15

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.0418. Min Act: -0.1700

Data Index: 425910 (C4 (Web Text))

Max Activating Token Index: 639

Text #16

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.0418. Min Act: -0.1700

Data Index: 1604828 (Python Code)

Max Activating Token Index: 312

Text #17

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.0103. Min Act: -0.1700

Data Index: 451589 (C4 (Web Text))

Max Activating Token Index: 376

Text #18

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.0576. Min Act: -0.1700

Data Index: 971882 (C4 (Web Text))

Max Activating Token Index: 750

Text #19

Max Range: 3.3095. Min Range: -3.3095

Max Act: 3.0576. Min Act: -0.1700

Data Index: 1300916 (C4 (Web Text))

Max Activating Token Index: 755