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

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

Neuron 1017 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: 10.6092. Min Range: -10.6092

Max Act: 10.6092. Min Act: -0.1700

Data Index: 791532 (C4 (Web Text))

Max Activating Token Index: 30

Text #1

Max Range: 10.6092. Min Range: -10.6092

Max Act: 10.4217. Min Act: -0.1700

Data Index: 547100 (C4 (Web Text))

Max Activating Token Index: 94

Text #2

Max Range: 10.6092. Min Range: -10.6092

Max Act: 10.2342. Min Act: -0.1700

Data Index: 594194 (C4 (Web Text))

Max Activating Token Index: 343

Text #3

Max Range: 10.6092. Min Range: -10.6092

Max Act: 10.2342. Min Act: -0.1700

Data Index: 153942 (C4 (Web Text))

Max Activating Token Index: 381

Text #4

Max Range: 10.6092. Min Range: -10.6092

Max Act: 10.1717. Min Act: -0.1700

Data Index: 1121829 (C4 (Web Text))

Max Activating Token Index: 593

Text #5

Max Range: 10.6092. Min Range: -10.6092

Max Act: 9.7967. Min Act: -0.1700

Data Index: 1167912 (C4 (Web Text))

Max Activating Token Index: 44

Text #6

Max Range: 10.6092. Min Range: -10.6092

Max Act: 9.9217. Min Act: -0.1700

Data Index: 362283 (C4 (Web Text))

Max Activating Token Index: 536

Text #7

Max Range: 10.6092. Min Range: -10.6092

Max Act: 9.6717. Min Act: -0.1700

Data Index: 1117703 (C4 (Web Text))

Max Activating Token Index: 279

Text #8

Max Range: 10.6092. Min Range: -10.6092

Max Act: 9.4217. Min Act: -0.1700

Data Index: 1085364 (C4 (Web Text))

Max Activating Token Index: 350

Text #9

Max Range: 10.6092. Min Range: -10.6092

Max Act: 9.2967. Min Act: -0.1700

Data Index: 389016 (C4 (Web Text))

Max Activating Token Index: 993

Text #10

Max Range: 10.6092. Min Range: -10.6092

Max Act: 9.2967. Min Act: -0.1700

Data Index: 1166349 (C4 (Web Text))

Max Activating Token Index: 557

Text #11

Max Range: 10.6092. Min Range: -10.6092

Max Act: 9.2342. Min Act: -0.1700

Data Index: 437735 (C4 (Web Text))

Max Activating Token Index: 699

Text #12

Max Range: 10.6092. Min Range: -10.6092

Max Act: 9.1092. Min Act: -0.1700

Data Index: 111021 (C4 (Web Text))

Max Activating Token Index: 788

Text #13

Max Range: 10.6092. Min Range: -10.6092

Max Act: 9.1092. Min Act: -0.1700

Data Index: 1105090 (C4 (Web Text))

Max Activating Token Index: 926

Text #14

Max Range: 10.6092. Min Range: -10.6092

Max Act: 8.9217. Min Act: -0.1699

Data Index: 1069834 (C4 (Web Text))

Max Activating Token Index: 372

Text #15

Max Range: 10.6092. Min Range: -10.6092

Max Act: 9.1092. Min Act: -0.1691

Data Index: 809221 (C4 (Web Text))

Max Activating Token Index: 280

Text #16

Max Range: 10.6092. Min Range: -10.6092

Max Act: 8.7967. Min Act: -0.1700

Data Index: 710918 (C4 (Web Text))

Max Activating Token Index: 192

Text #17

Max Range: 10.6092. Min Range: -10.6092

Max Act: 8.8592. Min Act: -0.1700

Data Index: 806830 (C4 (Web Text))

Max Activating Token Index: 422

Text #18

Max Range: 10.6092. Min Range: -10.6092

Max Act: 8.9842. Min Act: -0.1700

Data Index: 184813 (C4 (Web Text))

Max Activating Token Index: 785

Text #19

Max Range: 10.6092. Min Range: -10.6092

Max Act: 8.7967. Min Act: -0.1700

Data Index: 958273 (C4 (Web Text))

Max Activating Token Index: 562