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

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

Neuron 1784 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.2948. Min Act: -0.1700

Data Index: 986076 (C4 (Web Text))

Max Activating Token Index: 786

Text #1

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.3105. Min Act: -0.1700

Data Index: 642534 (C4 (Web Text))

Max Activating Token Index: 619

Text #2

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.2948. Min Act: -0.1700

Data Index: 949396 (C4 (Web Text))

Max Activating Token Index: 832

Text #3

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.2476. Min Act: -0.1700

Data Index: 589580 (C4 (Web Text))

Max Activating Token Index: 43

Text #4

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.2634. Min Act: -0.1700

Data Index: 29354 (C4 (Web Text))

Max Activating Token Index: 977

Text #5

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.1847. Min Act: -0.1700

Data Index: 1197997 (C4 (Web Text))

Max Activating Token Index: 507

Text #6

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.2005. Min Act: -0.1700

Data Index: 571216 (C4 (Web Text))

Max Activating Token Index: 164

Text #7

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.1690. Min Act: -0.1700

Data Index: 599970 (C4 (Web Text))

Max Activating Token Index: 152

Text #8

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.2005. Min Act: -0.1700

Data Index: 367648 (C4 (Web Text))

Max Activating Token Index: 462

Text #9

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.1690. Min Act: -0.1700

Data Index: 693318 (C4 (Web Text))

Max Activating Token Index: 663

Text #10

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.1690. Min Act: -0.1700

Data Index: 1158576 (C4 (Web Text))

Max Activating Token Index: 819

Text #11

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.1532. Min Act: -0.1700

Data Index: 981445 (C4 (Web Text))

Max Activating Token Index: 285

Text #12

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.1532. Min Act: -0.1700

Data Index: 339373 (C4 (Web Text))

Max Activating Token Index: 35

Text #13

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.1217. Min Act: -0.1700

Data Index: 20122 (C4 (Web Text))

Max Activating Token Index: 924

Text #14

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.1375. Min Act: -0.1700

Data Index: 325450 (C4 (Web Text))

Max Activating Token Index: 522

Text #15

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.1375. Min Act: -0.1700

Data Index: 87165 (C4 (Web Text))

Max Activating Token Index: 829

Text #16

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.1217. Min Act: -0.1700

Data Index: 220802 (C4 (Web Text))

Max Activating Token Index: 410

Text #17

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.1060. Min Act: -0.1700

Data Index: 918710 (C4 (Web Text))

Max Activating Token Index: 251

Text #18

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.0902. Min Act: -0.1700

Data Index: 1216407 (C4 (Web Text))

Max Activating Token Index: 996

Text #19

Max Range: 3.3105. Min Range: -3.3105

Max Act: 3.1217. Min Act: -0.1700

Data Index: 368715 (C4 (Web Text))

Max Activating Token Index: 669