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

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

Neuron 1579 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: 3.8419. Min Range: -3.8419

Max Act: 3.8419. Min Act: -0.1700

Data Index: 623111 (C4 (Web Text))

Max Activating Token Index: 900

Text #1

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.8106. Min Act: -0.1700

Data Index: 1446667 (Python Code)

Max Activating Token Index: 538

Text #2

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.6855. Min Act: -0.1700

Data Index: 1591365 (Python Code)

Max Activating Token Index: 540

Text #3

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.7168. Min Act: -0.1700

Data Index: 569663 (C4 (Web Text))

Max Activating Token Index: 627

Text #4

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.6855. Min Act: -0.1700

Data Index: 1616321 (Python Code)

Max Activating Token Index: 709

Text #5

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.7168. Min Act: -0.1700

Data Index: 1540044 (Python Code)

Max Activating Token Index: 431

Text #6

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.6542. Min Act: -0.1700

Data Index: 1479016 (Python Code)

Max Activating Token Index: 474

Text #7

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.6542. Min Act: -0.1700

Data Index: 784684 (C4 (Web Text))

Max Activating Token Index: 318

Text #8

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.5916. Min Act: -0.1700

Data Index: 928146 (C4 (Web Text))

Max Activating Token Index: 674

Text #9

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.6855. Min Act: -0.1700

Data Index: 646651 (C4 (Web Text))

Max Activating Token Index: 728

Text #10

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.6542. Min Act: -0.1700

Data Index: 431809 (C4 (Web Text))

Max Activating Token Index: 774

Text #11

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.6542. Min Act: -0.1700

Data Index: 348968 (C4 (Web Text))

Max Activating Token Index: 196

Text #12

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.6229. Min Act: -0.1700

Data Index: 1626028 (Python Code)

Max Activating Token Index: 616

Text #13

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.5916. Min Act: -0.1700

Data Index: 1390150 (Python Code)

Max Activating Token Index: 75

Text #14

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.5916. Min Act: -0.1700

Data Index: 1272133 (C4 (Web Text))

Max Activating Token Index: 936

Text #15

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.5602. Min Act: -0.1700

Data Index: 1548186 (Python Code)

Max Activating Token Index: 738

Text #16

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.5289. Min Act: -0.1700

Data Index: 1481133 (Python Code)

Max Activating Token Index: 18

Text #17

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.5602. Min Act: -0.1700

Data Index: 1047914 (C4 (Web Text))

Max Activating Token Index: 162

Text #18

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.5289. Min Act: -0.1700

Data Index: 320369 (C4 (Web Text))

Max Activating Token Index: 876

Text #19

Max Range: 3.8419. Min Range: -3.8419

Max Act: 3.5602. Min Act: -0.1700

Data Index: 21941 (C4 (Web Text))

Max Activating Token Index: 588