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

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

Neuron 1269 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: 5.3168. Min Range: -5.3168

Max Act: 5.3168. Min Act: -0.1700

Data Index: 131582 (C4 (Web Text))

Max Activating Token Index: 490

Text #1

Max Range: 5.3168. Min Range: -5.3168

Max Act: 5.0668. Min Act: -0.1700

Data Index: 853933 (C4 (Web Text))

Max Activating Token Index: 803

Text #2

Max Range: 5.3168. Min Range: -5.3168

Max Act: 5.0668. Min Act: -0.1700

Data Index: 135672 (C4 (Web Text))

Max Activating Token Index: 307

Text #3

Max Range: 5.3168. Min Range: -5.3168

Max Act: 5.0355. Min Act: -0.1699

Data Index: 1192845 (C4 (Web Text))

Max Activating Token Index: 334

Text #4

Max Range: 5.3168. Min Range: -5.3168

Max Act: 5.0043. Min Act: -0.1700

Data Index: 1024700 (C4 (Web Text))

Max Activating Token Index: 760

Text #5

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.9730. Min Act: -0.1700

Data Index: 11674 (C4 (Web Text))

Max Activating Token Index: 653

Text #6

Max Range: 5.3168. Min Range: -5.3168

Max Act: 5.0355. Min Act: -0.1700

Data Index: 1075146 (C4 (Web Text))

Max Activating Token Index: 432

Text #7

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.9730. Min Act: -0.1700

Data Index: 1114375 (C4 (Web Text))

Max Activating Token Index: 723

Text #8

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.9418. Min Act: -0.1700

Data Index: 1189123 (C4 (Web Text))

Max Activating Token Index: 736

Text #9

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.9105. Min Act: -0.1700

Data Index: 971982 (C4 (Web Text))

Max Activating Token Index: 381

Text #10

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.8793. Min Act: -0.1700

Data Index: 1053656 (C4 (Web Text))

Max Activating Token Index: 836

Text #11

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.8168. Min Act: -0.1700

Data Index: 429008 (C4 (Web Text))

Max Activating Token Index: 939

Text #12

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.8168. Min Act: -0.1700

Data Index: 620800 (C4 (Web Text))

Max Activating Token Index: 121

Text #13

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.8168. Min Act: -0.1700

Data Index: 1276409 (C4 (Web Text))

Max Activating Token Index: 891

Text #14

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.7855. Min Act: -0.1700

Data Index: 930394 (C4 (Web Text))

Max Activating Token Index: 1018

Text #15

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.8168. Min Act: -0.1700

Data Index: 908429 (C4 (Web Text))

Max Activating Token Index: 565

Text #16

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.7855. Min Act: -0.1700

Data Index: 482639 (C4 (Web Text))

Max Activating Token Index: 633

Text #17

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.7230. Min Act: -0.1700

Data Index: 969975 (C4 (Web Text))

Max Activating Token Index: 353

Text #18

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.7230. Min Act: -0.1700

Data Index: 80243 (C4 (Web Text))

Max Activating Token Index: 146

Text #19

Max Range: 5.3168. Min Range: -5.3168

Max Act: 4.7230. Min Act: -0.1700

Data Index: 343769 (C4 (Web Text))

Max Activating Token Index: 118