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

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

Neuron 777 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.0116. Min Range: -3.0116

Max Act: 3.0116. Min Act: -0.1700

Data Index: 1608087 (Python Code)

Max Activating Token Index: 380

Text #1

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.9958. Min Act: -0.1700

Data Index: 949537 (C4 (Web Text))

Max Activating Token Index: 611

Text #2

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.9641. Min Act: -0.1700

Data Index: 1219043 (C4 (Web Text))

Max Activating Token Index: 739

Text #3

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.9325. Min Act: -0.1700

Data Index: 298210 (C4 (Web Text))

Max Activating Token Index: 144

Text #4

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.9325. Min Act: -0.1700

Data Index: 313688 (C4 (Web Text))

Max Activating Token Index: 50

Text #5

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.9166. Min Act: -0.1700

Data Index: 1167761 (C4 (Web Text))

Max Activating Token Index: 493

Text #6

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.9007. Min Act: -0.1700

Data Index: 1471375 (Python Code)

Max Activating Token Index: 786

Text #7

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.9166. Min Act: -0.1700

Data Index: 987082 (C4 (Web Text))

Max Activating Token Index: 237

Text #8

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.8531. Min Act: -0.1700

Data Index: 1260456 (C4 (Web Text))

Max Activating Token Index: 982

Text #9

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.8849. Min Act: -0.1700

Data Index: 1276965 (C4 (Web Text))

Max Activating Token Index: 693

Text #10

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.8531. Min Act: -0.1700

Data Index: 556335 (C4 (Web Text))

Max Activating Token Index: 822

Text #11

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.8372. Min Act: -0.1700

Data Index: 409586 (C4 (Web Text))

Max Activating Token Index: 394

Text #12

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.8372. Min Act: -0.1700

Data Index: 372486 (C4 (Web Text))

Max Activating Token Index: 758

Text #13

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.8372. Min Act: -0.1700

Data Index: 1170236 (C4 (Web Text))

Max Activating Token Index: 88

Text #14

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.8849. Min Act: -0.1700

Data Index: 1368786 (Python Code)

Max Activating Token Index: 299

Text #15

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.8690. Min Act: -0.1700

Data Index: 1415807 (Python Code)

Max Activating Token Index: 790

Text #16

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.8531. Min Act: -0.1700

Data Index: 500714 (C4 (Web Text))

Max Activating Token Index: 125

Text #17

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.8531. Min Act: -0.1700

Data Index: 1144562 (C4 (Web Text))

Max Activating Token Index: 940

Text #18

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.8531. Min Act: -0.1700

Data Index: 1139848 (C4 (Web Text))

Max Activating Token Index: 733

Text #19

Max Range: 3.0116. Min Range: -3.0116

Max Act: 2.8054. Min Act: -0.1700

Data Index: 399738 (C4 (Web Text))

Max Activating Token Index: 937