Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 140 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.6991. Min Act: -0.1700

Data Index: 536883 (C4 (Web Text))

Max Activating Token Index: 76

Text #1

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.6678. Min Act: -0.1700

Data Index: 512376 (C4 (Web Text))

Max Activating Token Index: 703

Text #2

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.6366. Min Act: -0.1700

Data Index: 255210 (C4 (Web Text))

Max Activating Token Index: 46

Text #3

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.5428. Min Act: -0.1697

Data Index: 934729 (C4 (Web Text))

Max Activating Token Index: 546

Text #4

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.5741. Min Act: -0.1698

Data Index: 229709 (C4 (Web Text))

Max Activating Token Index: 98

Text #5

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.5428. Min Act: -0.1697

Data Index: 1107007 (C4 (Web Text))

Max Activating Token Index: 148

Text #6

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.4491. Min Act: -0.1700

Data Index: 427437 (C4 (Web Text))

Max Activating Token Index: 514

Text #7

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.4803. Min Act: -0.1700

Data Index: 988030 (C4 (Web Text))

Max Activating Token Index: 566

Text #8

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.4803. Min Act: -0.1700

Data Index: 307335 (C4 (Web Text))

Max Activating Token Index: 114

Text #9

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.4491. Min Act: -0.1700

Data Index: 823011 (C4 (Web Text))

Max Activating Token Index: 220

Text #10

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.4178. Min Act: -0.1697

Data Index: 1328711 (C4 (Web Text))

Max Activating Token Index: 263

Text #11

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.4178. Min Act: -0.1697

Data Index: 653979 (C4 (Web Text))

Max Activating Token Index: 228

Text #12

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.4178. Min Act: -0.1699

Data Index: 1205261 (C4 (Web Text))

Max Activating Token Index: 256

Text #13

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.3553. Min Act: -0.1699

Data Index: 484051 (C4 (Web Text))

Max Activating Token Index: 938

Text #14

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.3866. Min Act: -0.1700

Data Index: 329666 (C4 (Web Text))

Max Activating Token Index: 200

Text #15

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.3866. Min Act: -0.1700

Data Index: 1179353 (C4 (Web Text))

Max Activating Token Index: 415

Text #16

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.3866. Min Act: -0.1700

Data Index: 321777 (C4 (Web Text))

Max Activating Token Index: 467

Text #17

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.4178. Min Act: -0.1700

Data Index: 1076942 (C4 (Web Text))

Max Activating Token Index: 203

Text #18

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.3553. Min Act: -0.1700

Data Index: 1056684 (C4 (Web Text))

Max Activating Token Index: 508

Text #19

Max Range: 5.6991. Min Range: -5.6991

Max Act: 5.3866. Min Act: -0.1697

Data Index: 670941 (C4 (Web Text))

Max Activating Token Index: 179