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

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

Neuron 308 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.8856. Min Act: -0.1700

Data Index: 81364 (C4 (Web Text))

Max Activating Token Index: 1007

Text #1

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.8231. Min Act: -0.1700

Data Index: 691292 (C4 (Web Text))

Max Activating Token Index: 220

Text #2

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.8544. Min Act: -0.1700

Data Index: 786870 (C4 (Web Text))

Max Activating Token Index: 701

Text #3

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.8231. Min Act: -0.1700

Data Index: 49634 (C4 (Web Text))

Max Activating Token Index: 174

Text #4

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.8231. Min Act: -0.1700

Data Index: 857717 (C4 (Web Text))

Max Activating Token Index: 976

Text #5

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.8231. Min Act: -0.1700

Data Index: 777402 (C4 (Web Text))

Max Activating Token Index: 918

Text #6

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.8231. Min Act: -0.1700

Data Index: 1299589 (C4 (Web Text))

Max Activating Token Index: 859

Text #7

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.7918. Min Act: -0.1700

Data Index: 353850 (C4 (Web Text))

Max Activating Token Index: 217

Text #8

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.7292. Min Act: -0.1700

Data Index: 1292693 (C4 (Web Text))

Max Activating Token Index: 207

Text #9

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.7292. Min Act: -0.1700

Data Index: 763901 (C4 (Web Text))

Max Activating Token Index: 784

Text #10

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.7605. Min Act: -0.1700

Data Index: 379677 (C4 (Web Text))

Max Activating Token Index: 449

Text #11

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.7605. Min Act: -0.1700

Data Index: 300644 (C4 (Web Text))

Max Activating Token Index: 749

Text #12

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.7292. Min Act: -0.1700

Data Index: 350900 (C4 (Web Text))

Max Activating Token Index: 550

Text #13

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.7292. Min Act: -0.1700

Data Index: 80624 (C4 (Web Text))

Max Activating Token Index: 1005

Text #14

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.6979. Min Act: -0.1700

Data Index: 1265605 (C4 (Web Text))

Max Activating Token Index: 545

Text #15

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.7292. Min Act: -0.1700

Data Index: 1234924 (C4 (Web Text))

Max Activating Token Index: 451

Text #16

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.6979. Min Act: -0.1700

Data Index: 1109877 (C4 (Web Text))

Max Activating Token Index: 356

Text #17

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.6823. Min Act: -0.1700

Data Index: 372542 (C4 (Web Text))

Max Activating Token Index: 90

Text #18

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.6353. Min Act: -0.1700

Data Index: 798981 (C4 (Web Text))

Max Activating Token Index: 867

Text #19

Max Range: 3.8856. Min Range: -3.8856

Max Act: 3.6353. Min Act: -0.1699

Data Index: 1051119 (C4 (Web Text))

Max Activating Token Index: 524