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

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

Neuron 1225 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.8421. Min Act: -0.1700

Data Index: 1005132 (C4 (Web Text))

Max Activating Token Index: 885

Text #1

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.8265. Min Act: -0.1700

Data Index: 299933 (C4 (Web Text))

Max Activating Token Index: 887

Text #2

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.8265. Min Act: -0.1700

Data Index: 1410691 (Python Code)

Max Activating Token Index: 1010

Text #3

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.8108. Min Act: -0.1700

Data Index: 248007 (C4 (Web Text))

Max Activating Token Index: 521

Text #4

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.8265. Min Act: -0.1700

Data Index: 862829 (C4 (Web Text))

Max Activating Token Index: 590

Text #5

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.7483. Min Act: -0.1700

Data Index: 328999 (C4 (Web Text))

Max Activating Token Index: 517

Text #6

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.7483. Min Act: -0.1700

Data Index: 1016917 (C4 (Web Text))

Max Activating Token Index: 269

Text #7

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.7326. Min Act: -0.1700

Data Index: 1329366 (C4 (Web Text))

Max Activating Token Index: 307

Text #8

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.7170. Min Act: -0.1700

Data Index: 125362 (C4 (Web Text))

Max Activating Token Index: 457

Text #9

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.7170. Min Act: -0.1700

Data Index: 461197 (C4 (Web Text))

Max Activating Token Index: 986

Text #10

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.7013. Min Act: -0.1700

Data Index: 269172 (C4 (Web Text))

Max Activating Token Index: 593

Text #11

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.7013. Min Act: -0.1700

Data Index: 1358832 (C4 (Web Text))

Max Activating Token Index: 593

Text #12

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.6857. Min Act: -0.1700

Data Index: 1116640 (C4 (Web Text))

Max Activating Token Index: 649

Text #13

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.7013. Min Act: -0.1700

Data Index: 868787 (C4 (Web Text))

Max Activating Token Index: 637

Text #14

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.6544. Min Act: -0.1700

Data Index: 1251549 (C4 (Web Text))

Max Activating Token Index: 390

Text #15

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.6700. Min Act: -0.1700

Data Index: 1315604 (C4 (Web Text))

Max Activating Token Index: 3

Text #16

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.6700. Min Act: -0.1700

Data Index: 556052 (C4 (Web Text))

Max Activating Token Index: 745

Text #17

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.6544. Min Act: -0.1700

Data Index: 872157 (C4 (Web Text))

Max Activating Token Index: 507

Text #18

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.6544. Min Act: -0.1700

Data Index: 805269 (C4 (Web Text))

Max Activating Token Index: 243

Text #19

Max Range: 3.8421. Min Range: -3.8421

Max Act: 3.6387. Min Act: -0.1700

Data Index: 1334700 (C4 (Web Text))

Max Activating Token Index: 723