Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 672 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.8363. Min Act: -0.1700

Data Index: 436304 (C4 (Web Text))

Max Activating Token Index: 588

Text #1

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.8051. Min Act: -0.1700

Data Index: 782097 (C4 (Web Text))

Max Activating Token Index: 435

Text #2

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.7426. Min Act: -0.1700

Data Index: 1233733 (C4 (Web Text))

Max Activating Token Index: 435

Text #3

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.6801. Min Act: -0.1700

Data Index: 1332088 (C4 (Web Text))

Max Activating Token Index: 396

Text #4

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.7738. Min Act: -0.1700

Data Index: 68656 (C4 (Web Text))

Max Activating Token Index: 513

Text #5

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.6176. Min Act: -0.1700

Data Index: 1073797 (C4 (Web Text))

Max Activating Token Index: 119

Text #6

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.5863. Min Act: -0.1700

Data Index: 1153143 (C4 (Web Text))

Max Activating Token Index: 335

Text #7

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.4613. Min Act: -0.1700

Data Index: 934625 (C4 (Web Text))

Max Activating Token Index: 807

Text #8

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.3988. Min Act: -0.1700

Data Index: 986562 (C4 (Web Text))

Max Activating Token Index: 93

Text #9

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.3051. Min Act: -0.1700

Data Index: 72209 (C4 (Web Text))

Max Activating Token Index: 758

Text #10

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.1801. Min Act: -0.1700

Data Index: 564758 (C4 (Web Text))

Max Activating Token Index: 602

Text #11

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.1488. Min Act: -0.1700

Data Index: 442043 (C4 (Web Text))

Max Activating Token Index: 802

Text #12

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.0863. Min Act: -0.1700

Data Index: 546055 (C4 (Web Text))

Max Activating Token Index: 686

Text #13

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.1176. Min Act: -0.1700

Data Index: 801284 (C4 (Web Text))

Max Activating Token Index: 513

Text #14

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.1176. Min Act: -0.1700

Data Index: 852530 (C4 (Web Text))

Max Activating Token Index: 171

Text #15

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.0863. Min Act: -0.1700

Data Index: 1341039 (C4 (Web Text))

Max Activating Token Index: 790

Text #16

Max Range: 5.8363. Min Range: -5.8363

Max Act: 4.9926. Min Act: -0.1700

Data Index: 1249163 (C4 (Web Text))

Max Activating Token Index: 956

Text #17

Max Range: 5.8363. Min Range: -5.8363

Max Act: 5.0551. Min Act: -0.1700

Data Index: 788541 (C4 (Web Text))

Max Activating Token Index: 585

Text #18

Max Range: 5.8363. Min Range: -5.8363

Max Act: 4.9926. Min Act: -0.1700

Data Index: 1343031 (C4 (Web Text))

Max Activating Token Index: 624

Text #19

Max Range: 5.8363. Min Range: -5.8363

Max Act: 4.9926. Min Act: -0.1700

Data Index: 1121681 (C4 (Web Text))

Max Activating Token Index: 316