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

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

Neuron 1551 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.4907. Min Act: -0.1700

Data Index: 364419 (C4 (Web Text))

Max Activating Token Index: 932

Text #1

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.4282. Min Act: -0.1700

Data Index: 941420 (C4 (Web Text))

Max Activating Token Index: 879

Text #2

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.4282. Min Act: -0.1699

Data Index: 32064 (C4 (Web Text))

Max Activating Token Index: 213

Text #3

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.3032. Min Act: -0.1700

Data Index: 679933 (C4 (Web Text))

Max Activating Token Index: 156

Text #4

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.2719. Min Act: -0.1700

Data Index: 112976 (C4 (Web Text))

Max Activating Token Index: 28

Text #5

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.1782. Min Act: -0.1700

Data Index: 769738 (C4 (Web Text))

Max Activating Token Index: 141

Text #6

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.2094. Min Act: -0.1700

Data Index: 708448 (C4 (Web Text))

Max Activating Token Index: 119

Text #7

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.2094. Min Act: -0.1700

Data Index: 484051 (C4 (Web Text))

Max Activating Token Index: 876

Text #8

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.1782. Min Act: -0.1700

Data Index: 1000408 (C4 (Web Text))

Max Activating Token Index: 290

Text #9

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.1782. Min Act: -0.1699

Data Index: 1165612 (C4 (Web Text))

Max Activating Token Index: 926

Text #10

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.1157. Min Act: -0.1700

Data Index: 850489 (C4 (Web Text))

Max Activating Token Index: 631

Text #11

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.1782. Min Act: -0.1700

Data Index: 1062580 (C4 (Web Text))

Max Activating Token Index: 300

Text #12

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.0844. Min Act: -0.1700

Data Index: 1092344 (C4 (Web Text))

Max Activating Token Index: 211

Text #13

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.1157. Min Act: -0.1699

Data Index: 1122912 (C4 (Web Text))

Max Activating Token Index: 748

Text #14

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.1157. Min Act: -0.1700

Data Index: 259004 (C4 (Web Text))

Max Activating Token Index: 545

Text #15

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.0532. Min Act: -0.1683

Data Index: 407841 (C4 (Web Text))

Max Activating Token Index: 316

Text #16

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.0532. Min Act: -0.1700

Data Index: 1121965 (C4 (Web Text))

Max Activating Token Index: 641

Text #17

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.1157. Min Act: -0.1700

Data Index: 71625 (C4 (Web Text))

Max Activating Token Index: 817

Text #18

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.0532. Min Act: -0.1700

Data Index: 1131746 (C4 (Web Text))

Max Activating Token Index: 766

Text #19

Max Range: 5.4907. Min Range: -5.4907

Max Act: 5.0219. Min Act: -0.1695

Data Index: 278607 (C4 (Web Text))

Max Activating Token Index: 219