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

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

Neuron 313 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: 5.1703. Min Range: -5.1703

Max Act: 5.0765. Min Act: -0.1700

Data Index: 1399400 (Python Code)

Max Activating Token Index: 675

Text #1

Max Range: 5.1703. Min Range: -5.1703

Max Act: 5.1703. Min Act: -0.1700

Data Index: 1000349 (C4 (Web Text))

Max Activating Token Index: 299

Text #2

Max Range: 5.1703. Min Range: -5.1703

Max Act: 5.1703. Min Act: -0.1700

Data Index: 1517122 (Python Code)

Max Activating Token Index: 774

Text #3

Max Range: 5.1703. Min Range: -5.1703

Max Act: 5.1390. Min Act: -0.1700

Data Index: 1135278 (C4 (Web Text))

Max Activating Token Index: 962

Text #4

Max Range: 5.1703. Min Range: -5.1703

Max Act: 5.1390. Min Act: -0.1700

Data Index: 607040 (C4 (Web Text))

Max Activating Token Index: 124

Text #5

Max Range: 5.1703. Min Range: -5.1703

Max Act: 5.0765. Min Act: -0.1700

Data Index: 598493 (C4 (Web Text))

Max Activating Token Index: 471

Text #6

Max Range: 5.1703. Min Range: -5.1703

Max Act: 5.0296. Min Act: -0.1700

Data Index: 1599025 (Python Code)

Max Activating Token Index: 720

Text #7

Max Range: 5.1703. Min Range: -5.1703

Max Act: 5.0609. Min Act: -0.1700

Data Index: 619865 (C4 (Web Text))

Max Activating Token Index: 427

Text #8

Max Range: 5.1703. Min Range: -5.1703

Max Act: 5.0296. Min Act: -0.1699

Data Index: 22470 (C4 (Web Text))

Max Activating Token Index: 319

Text #9

Max Range: 5.1703. Min Range: -5.1703

Max Act: 5.0296. Min Act: -0.1700

Data Index: 1266827 (C4 (Web Text))

Max Activating Token Index: 652

Text #10

Max Range: 5.1703. Min Range: -5.1703

Max Act: 4.9828. Min Act: -0.1700

Data Index: 102876 (C4 (Web Text))

Max Activating Token Index: 874

Text #11

Max Range: 5.1703. Min Range: -5.1703

Max Act: 4.9671. Min Act: -0.1700

Data Index: 362056 (C4 (Web Text))

Max Activating Token Index: 671

Text #12

Max Range: 5.1703. Min Range: -5.1703

Max Act: 5.0140. Min Act: -0.1700

Data Index: 572922 (C4 (Web Text))

Max Activating Token Index: 285

Text #13

Max Range: 5.1703. Min Range: -5.1703

Max Act: 4.9828. Min Act: -0.1700

Data Index: 679152 (C4 (Web Text))

Max Activating Token Index: 100

Text #14

Max Range: 5.1703. Min Range: -5.1703

Max Act: 5.0296. Min Act: -0.1700

Data Index: 971684 (C4 (Web Text))

Max Activating Token Index: 838

Text #15

Max Range: 5.1703. Min Range: -5.1703

Max Act: 4.9515. Min Act: -0.1700

Data Index: 1563853 (Python Code)

Max Activating Token Index: 720

Text #16

Max Range: 5.1703. Min Range: -5.1703

Max Act: 5.0140. Min Act: -0.1700

Data Index: 809781 (C4 (Web Text))

Max Activating Token Index: 318

Text #17

Max Range: 5.1703. Min Range: -5.1703

Max Act: 4.9671. Min Act: -0.1700

Data Index: 594937 (C4 (Web Text))

Max Activating Token Index: 107

Text #18

Max Range: 5.1703. Min Range: -5.1703

Max Act: 4.9515. Min Act: -0.1700

Data Index: 1391838 (Python Code)

Max Activating Token Index: 579

Text #19

Max Range: 5.1703. Min Range: -5.1703

Max Act: 4.9515. Min Act: -0.1700

Data Index: 919221 (C4 (Web Text))

Max Activating Token Index: 177