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

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

Neuron 626 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.2703. Min Act: -0.1700

Data Index: 1652487 (Python Code)

Max Activating Token Index: 154

Text #1

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.1917. Min Act: -0.1700

Data Index: 1368160 (Python Code)

Max Activating Token Index: 987

Text #2

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.1445. Min Act: -0.1700

Data Index: 1535087 (Python Code)

Max Activating Token Index: 55

Text #3

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0972. Min Act: -0.1700

Data Index: 1375735 (Python Code)

Max Activating Token Index: 467

Text #4

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0814. Min Act: -0.1700

Data Index: 1554719 (Python Code)

Max Activating Token Index: 299

Text #5

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0814. Min Act: -0.1700

Data Index: 912489 (C4 (Web Text))

Max Activating Token Index: 966

Text #6

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0656. Min Act: -0.1700

Data Index: 906009 (C4 (Web Text))

Max Activating Token Index: 986

Text #7

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0814. Min Act: -0.1700

Data Index: 1380974 (Python Code)

Max Activating Token Index: 682

Text #8

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0656. Min Act: -0.1700

Data Index: 1600797 (Python Code)

Max Activating Token Index: 38

Text #9

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0025. Min Act: -0.1700

Data Index: 174577 (C4 (Web Text))

Max Activating Token Index: 421

Text #10

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0814. Min Act: -0.1700

Data Index: 414666 (C4 (Web Text))

Max Activating Token Index: 432

Text #11

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0183. Min Act: -0.1700

Data Index: 1402170 (Python Code)

Max Activating Token Index: 842

Text #12

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0341. Min Act: -0.1700

Data Index: 1547578 (Python Code)

Max Activating Token Index: 369

Text #13

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0025. Min Act: -0.1700

Data Index: 447324 (C4 (Web Text))

Max Activating Token Index: 302

Text #14

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0025. Min Act: -0.1700

Data Index: 342570 (C4 (Web Text))

Max Activating Token Index: 483

Text #15

Max Range: 3.2703. Min Range: -3.2703

Max Act: 2.9866. Min Act: -0.1700

Data Index: 909060 (C4 (Web Text))

Max Activating Token Index: 345

Text #16

Max Range: 3.2703. Min Range: -3.2703

Max Act: 2.9866. Min Act: -0.1700

Data Index: 277192 (C4 (Web Text))

Max Activating Token Index: 501

Text #17

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0341. Min Act: -0.1700

Data Index: 1469324 (Python Code)

Max Activating Token Index: 901

Text #18

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0025. Min Act: -0.1700

Data Index: 1557498 (Python Code)

Max Activating Token Index: 985

Text #19

Max Range: 3.2703. Min Range: -3.2703

Max Act: 3.0025. Min Act: -0.1700

Data Index: 1384131 (Python Code)

Max Activating Token Index: 240