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

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

Neuron 26 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: 3.8271. Min Range: -3.8271

Max Act: 3.8271. Min Act: -0.1700

Data Index: 646879 (C4 (Web Text))

Max Activating Token Index: 74

Text #1

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.7802. Min Act: -0.1700

Data Index: 802127 (C4 (Web Text))

Max Activating Token Index: 538

Text #2

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.7802. Min Act: -0.1700

Data Index: 269789 (C4 (Web Text))

Max Activating Token Index: 31

Text #3

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.7802. Min Act: -0.1700

Data Index: 397327 (C4 (Web Text))

Max Activating Token Index: 157

Text #4

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6863. Min Act: -0.1700

Data Index: 861066 (C4 (Web Text))

Max Activating Token Index: 20

Text #5

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.7019. Min Act: -0.1700

Data Index: 306821 (C4 (Web Text))

Max Activating Token Index: 97

Text #6

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6706. Min Act: -0.1700

Data Index: 689985 (C4 (Web Text))

Max Activating Token Index: 87

Text #7

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6550. Min Act: -0.1700

Data Index: 985666 (C4 (Web Text))

Max Activating Token Index: 67

Text #8

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6706. Min Act: -0.1700

Data Index: 733449 (C4 (Web Text))

Max Activating Token Index: 765

Text #9

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6706. Min Act: -0.1700

Data Index: 1132278 (C4 (Web Text))

Max Activating Token Index: 806

Text #10

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6550. Min Act: -0.1700

Data Index: 63187 (C4 (Web Text))

Max Activating Token Index: 419

Text #11

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6550. Min Act: -0.1700

Data Index: 816312 (C4 (Web Text))

Max Activating Token Index: 975

Text #12

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6393. Min Act: -0.1700

Data Index: 458615 (C4 (Web Text))

Max Activating Token Index: 167

Text #13

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6393. Min Act: -0.1700

Data Index: 1145875 (C4 (Web Text))

Max Activating Token Index: 210

Text #14

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6237. Min Act: -0.1700

Data Index: 1138943 (C4 (Web Text))

Max Activating Token Index: 85

Text #15

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6393. Min Act: -0.1700

Data Index: 910852 (C4 (Web Text))

Max Activating Token Index: 269

Text #16

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6080. Min Act: -0.1700

Data Index: 1301521 (C4 (Web Text))

Max Activating Token Index: 16

Text #17

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6237. Min Act: -0.1700

Data Index: 1341694 (C4 (Web Text))

Max Activating Token Index: 472

Text #18

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6237. Min Act: -0.1700

Data Index: 639053 (C4 (Web Text))

Max Activating Token Index: 205

Text #19

Max Range: 3.8271. Min Range: -3.8271

Max Act: 3.6393. Min Act: -0.1700

Data Index: 878471 (C4 (Web Text))

Max Activating Token Index: 46