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

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

Neuron 1617 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: 5.5420. Min Range: -5.5420

Max Act: 5.5420. Min Act: -0.1700

Data Index: 394721 (C4 (Web Text))

Max Activating Token Index: 155

Text #1

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.5107. Min Act: -0.1700

Data Index: 1366585 (Python Code)

Max Activating Token Index: 860

Text #2

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.4795. Min Act: -0.1700

Data Index: 1422918 (Python Code)

Max Activating Token Index: 856

Text #3

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.4170. Min Act: -0.1700

Data Index: 1625222 (Python Code)

Max Activating Token Index: 154

Text #4

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.2920. Min Act: -0.1700

Data Index: 1564044 (Python Code)

Max Activating Token Index: 848

Text #5

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.2607. Min Act: -0.1700

Data Index: 334109 (C4 (Web Text))

Max Activating Token Index: 445

Text #6

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.2920. Min Act: -0.1700

Data Index: 1444491 (Python Code)

Max Activating Token Index: 955

Text #7

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.2295. Min Act: -0.1700

Data Index: 1236333 (C4 (Web Text))

Max Activating Token Index: 157

Text #8

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.2295. Min Act: -0.1700

Data Index: 1106196 (C4 (Web Text))

Max Activating Token Index: 114

Text #9

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.2295. Min Act: -0.1700

Data Index: 844628 (C4 (Web Text))

Max Activating Token Index: 19

Text #10

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.1982. Min Act: -0.1700

Data Index: 1619094 (Python Code)

Max Activating Token Index: 237

Text #11

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.1670. Min Act: -0.1700

Data Index: 1573044 (Python Code)

Max Activating Token Index: 336

Text #12

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.1357. Min Act: -0.1700

Data Index: 1548768 (Python Code)

Max Activating Token Index: 319

Text #13

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.1670. Min Act: -0.1700

Data Index: 1437191 (Python Code)

Max Activating Token Index: 817

Text #14

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.1357. Min Act: -0.1700

Data Index: 1366054 (Python Code)

Max Activating Token Index: 331

Text #15

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.0107. Min Act: -0.1700

Data Index: 61425 (C4 (Web Text))

Max Activating Token Index: 860

Text #16

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.0732. Min Act: -0.1700

Data Index: 1586069 (Python Code)

Max Activating Token Index: 609

Text #17

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.0732. Min Act: -0.1700

Data Index: 1481471 (Python Code)

Max Activating Token Index: 692

Text #18

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.0420. Min Act: -0.1700

Data Index: 1598852 (Python Code)

Max Activating Token Index: 971

Text #19

Max Range: 5.5420. Min Range: -5.5420

Max Act: 5.0420. Min Act: -0.1700

Data Index: 739297 (C4 (Web Text))

Max Activating Token Index: 740