Model: GELU Model: 1 Layers, 2048 Neurons per Layer

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

Neuron 12 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('gelu-1l')

Click the toggle to see the full text!

Text #0

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.6015. Min Act: -0.1697

Data Index: 1429167 (Python Code)

Max Activating Token Index: 728

Text #1

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.5373. Min Act: -0.1700

Data Index: 1380476 (Python Code)

Max Activating Token Index: 975

Text #2

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.3956. Min Act: -0.1700

Data Index: 1008779 (C4 (Web Text))

Max Activating Token Index: 932

Text #3

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.4121. Min Act: -0.1699

Data Index: 361069 (C4 (Web Text))

Max Activating Token Index: 179

Text #4

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.3654. Min Act: -0.1700

Data Index: 174230 (C4 (Web Text))

Max Activating Token Index: 285

Text #5

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.2949. Min Act: -0.1700

Data Index: 81746 (C4 (Web Text))

Max Activating Token Index: 504

Text #6

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.2607. Min Act: -0.1700

Data Index: 815867 (C4 (Web Text))

Max Activating Token Index: 664

Text #7

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.2326. Min Act: -0.1699

Data Index: 1566046 (Python Code)

Max Activating Token Index: 201

Text #8

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.2177. Min Act: -0.1700

Data Index: 1548690 (Python Code)

Max Activating Token Index: 875

Text #9

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.2125. Min Act: -0.1636

Data Index: 1561088 (Python Code)

Max Activating Token Index: 292

Text #10

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.2066. Min Act: -0.1700

Data Index: 1099671 (C4 (Web Text))

Max Activating Token Index: 590

Text #11

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.1690. Min Act: -0.1700

Data Index: 1581798 (Python Code)

Max Activating Token Index: 847

Text #12

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.1663. Min Act: -0.1700

Data Index: 1428462 (Python Code)

Max Activating Token Index: 900

Text #13

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.1694. Min Act: -0.1697

Data Index: 1641331 (Python Code)

Max Activating Token Index: 909

Text #14

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.0783. Min Act: -0.1699

Data Index: 740226 (C4 (Web Text))

Max Activating Token Index: 627

Text #15

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.0493. Min Act: -0.1699

Data Index: 28081 (C4 (Web Text))

Max Activating Token Index: 502

Text #16

Max Range: 4.6015. Min Range: -4.6015

Max Act: 4.0020. Min Act: -0.1700

Data Index: 955819 (C4 (Web Text))

Max Activating Token Index: 121

Text #17

Max Range: 4.6015. Min Range: -4.6015

Max Act: 3.9842. Min Act: -0.1700

Data Index: 951384 (C4 (Web Text))

Max Activating Token Index: 250

Text #18

Max Range: 4.6015. Min Range: -4.6015

Max Act: 3.9924. Min Act: -0.1700

Data Index: 554966 (C4 (Web Text))

Max Activating Token Index: 122

Text #19

Max Range: 4.6015. Min Range: -4.6015

Max Act: 3.9941. Min Act: -0.1700

Data Index: 68184 (C4 (Web Text))

Max Activating Token Index: 920