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

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

Neuron 374 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2291. Min Range: -5.2291

Max Act: 5.2291. Min Act: -0.1700

Data Index: 1387160 (Python Code)

Max Activating Token Index: 698

Text #1

Max Range: 5.2291. Min Range: -5.2291

Max Act: 5.1041. Min Act: -0.1700

Data Index: 1391089 (Python Code)

Max Activating Token Index: 696

Text #2

Max Range: 5.2291. Min Range: -5.2291

Max Act: 5.0416. Min Act: -0.1700

Data Index: 372085 (C4 (Web Text))

Max Activating Token Index: 733

Text #3

Max Range: 5.2291. Min Range: -5.2291

Max Act: 5.0728. Min Act: -0.1700

Data Index: 1398091 (Python Code)

Max Activating Token Index: 574

Text #4

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.9791. Min Act: -0.1700

Data Index: 1554570 (Python Code)

Max Activating Token Index: 521

Text #5

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.9791. Min Act: -0.1700

Data Index: 1486668 (Python Code)

Max Activating Token Index: 326

Text #6

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.9166. Min Act: -0.1700

Data Index: 1405418 (Python Code)

Max Activating Token Index: 346

Text #7

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.8853. Min Act: -0.1700

Data Index: 1408695 (Python Code)

Max Activating Token Index: 592

Text #8

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.9166. Min Act: -0.1700

Data Index: 1626830 (Python Code)

Max Activating Token Index: 406

Text #9

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.9166. Min Act: -0.1700

Data Index: 702800 (C4 (Web Text))

Max Activating Token Index: 1019

Text #10

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.8853. Min Act: -0.1700

Data Index: 1556825 (Python Code)

Max Activating Token Index: 988

Text #11

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.8541. Min Act: -0.1700

Data Index: 902209 (C4 (Web Text))

Max Activating Token Index: 167

Text #12

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.8228. Min Act: -0.1700

Data Index: 1382925 (Python Code)

Max Activating Token Index: 198

Text #13

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.8853. Min Act: -0.1700

Data Index: 1440340 (Python Code)

Max Activating Token Index: 551

Text #14

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.7916. Min Act: -0.1700

Data Index: 1517319 (Python Code)

Max Activating Token Index: 894

Text #15

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.8853. Min Act: -0.1700

Data Index: 1388409 (Python Code)

Max Activating Token Index: 901

Text #16

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.7916. Min Act: -0.1700

Data Index: 1444955 (Python Code)

Max Activating Token Index: 748

Text #17

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.8228. Min Act: -0.1700

Data Index: 1388790 (Python Code)

Max Activating Token Index: 703

Text #18

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.7603. Min Act: -0.1700

Data Index: 45725 (C4 (Web Text))

Max Activating Token Index: 470

Text #19

Max Range: 5.2291. Min Range: -5.2291

Max Act: 4.7916. Min Act: -0.1700

Data Index: 994815 (C4 (Web Text))

Max Activating Token Index: 559